Full text of "Rocket propulsion elements"

See other formats

ROCKET

PROPULSION

ELEMENTS

Eighth Edition

GEORGE P. SUTTON
OSCAR BIBLARZ

ROCKET PROPULSION ELEMENTS

The book’s cover shows a typical separation of a payload stage or satellite from
an upper stage of a space flight vehicle in the vacuum of space. There are two
rocket propulsion systems. The liquid propellant rocket engine on the upper stage
(on the lower left) has just lifted the satellite into space and has just been shut
down. The reaction control system on the payload stage has four small inclined
thrusters, which can be seen during their firing operation; they are moving the
payload away from the expended upper stage. A small thruster’s plume (flame)
has a bright, elliptically shaped, hot gas, very small core, which is visible. The
hot exhaust gases disperse from this core in all directions, and this part of the
plume is almost invisible.

Rocket Propulsion Elements

Eighth Edition

GEORGE P. SUTTON

Consultant

Formerly Laboratory Associate

Lawrence Livermore National Laboratory and formerly

Executive Director, Engineering

Rocketdyne Division of The Boeing Company

Now Pratt & Whitney Rocketdyne

OSCAR BIBLARZ

Professor Emeritus

Department of Mechanicai and Astronauticai Engineering
Naval Postgraduate School

WILEY

JOHN WILEY & SONS, INC.

This book is printed on acid-free paper.®

Copyright © 2010 hy John Wiley & Sons, Inc. All rights reserved
Published by John Wiley & Sons, Inc., Hohoken, New Jersey
Published simultaneously in Canada

No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any
form or by any means, electronic, mechanical, photocopying, recording, scanning, or otherwise,
except as permitted under Section 107 or 108 of the 1976 United States Copyright Act, with-
out either the prior written permission of the Publisher, or authorization through payment of the
appropriate per-copy fee to the Copyright Clearance Center, 222 Rosewood Drive, Danvers, MA
01923, (978) 750-8400, fax (978) 646-8600, or on the web at www.copyright.com. Requests to
the Publisher for permission should be addressed to the Permissions Department, John Wiley &
Sons, Inc., Ill River Street, Hoboken, NJ 07030, (201) 748-6011, fax (201) 748-6008, or online at
www.wiley.com/go/permissions.

Limit of Liability/Disclaimer of Warranty: While the publisher and the authors have used their
best efforts in preparing this book, they make no representations or warranties with respect to the
accuracy or completeness of the contents of this book and specihcally disclaim any implied warranties
of merchantability or fitness for a particular purpose. No warranty may be created or extended by
sales representatives or written sales materials. The advice and strategies contained herein may not
be suitable for your situation. You should consult with a professional where appropriate. Neither the
publisher nor the authors shall be liable for any accidents or loss of proht or any other commercial
damages, including but not limited to special, incidental, consequential, or other damages.

For general information about our other products and services, please contact our Customer Care
Department within the United States at (800) 762-2974, outside the United States at (317) 572-3993
or fax (317) 572-4002.

Wiley also publishes its books in a variety of electronic formats. Some content that appears in print
may not be available in electronic books. For more information about Wiley products, visit our web
site at www.wiley.com.

Library of Congress Cataloging-in-Publication Data:

Sutton, George Paul.

Rocket propulsion elements / George P. Sutton, Oscar Biblarz. — 8th ed.
p. cm.

Includes bibliographical references and index.

ISBN 978-0-470-08024-5 (cloth)

1. Rocket engines. I. Biblarz, Oscar. II. Title.

TL782.S8 2010
629.47'5— dc22

2009042591

Printed in the United States of America

10 987654321

CONTENTS

PREFACE xiii

1 Classification 1

1.1. Duct Jet Propulsion / 2

1 .2. Rocket Propulsion / 4

1.3. Applications of Rocket Propulsion / 14
References / 26

2 Definitions and Fundamentals 28

2.1. Definitions / 28

2.2. Thrust / 33

2.3. Exhaust Velocity / 36

2.4. Energy and Efficiencies / 38

2.5. Multiple Propulsion Systems / 40

2.6. Typical Performance Values / 41
Problems / 43

Symbols / 45
References / 46

3 Nozzle Theory and Thermodynamic Relations 47

3.1. Ideal Rocket / 48

3.2. Summary of Thermodynamic Relations / 49

CONTENTS

3.3. Isentropic Flow Through Nozzles / 53

3.4. Nozzle Configurations / 75

3.5. Real Nozzles / 85

3.6. Nozzle Alignment / 96
Problems / 97
Symbols / 100
References / 101

4 Flight Performance 103

4.1. Gravity-Free Drag-Free Space Flight / 103

4.2. Forces Acting on a Vehicle in the Atmosphere / 107

4.3. Basic Relations of Motion / 110

4.4. Space Flight / 116

4.5. Flight Maneuvers / 131

4.6. Effect of Propulsion System on Vehicle Performance / 137

4.7. Flight Vehicles / 139

4.8. Military Missiles / 148

4.9. Flight Stability / 152
Problems / 153
Symbols / 155
References / 157

5 Chemical Rocket Propellant Performance Analysis 158

5.1. Background and Fundamentals / 159

5.2. Analysis of Chamber or Motor Case Conditions / 165

5.3. Analysis of Nozzle Expansion Processes / 170

5.4. Computer-Assisted Analysis / 175

5.5. Results of Thermochemical Calculations / 176
Problems / 190

Symbols / 191
References / 192

6 Liquid Propellant Rocket Engine Fundamentals 194

6.1. Types of Propellants / 198

6.2. Propellant Tanks / 201

6.3. Propellant Feed Systems / 208

6.4. Gas Pressure Feed Systems / 210

6.5. Tank Pressurization / 217

6.6. Turbopump Feed Systems and Engine Cycles / 221

CONTENTS Vii

6.7. Rocket Engines for Maneuvering, Orbit Adjustments, or Attitude

Control / 229

6.8. Engine Eamilies / 233

6.9. Valves and Pipelines / 236

6.10. Engine Support Structure / 239
Problems / 240

Symbols / 242
References / 243

7 Liquid Propellants 245

7.1. Propellant Properties / 246

7.2. Liquid Oxidizers / 256

7.3. Liquid Euels / 259

7.4. Liquid Monopropellants / 263

7.5. Gelled Propellants / 265

7.6. Gaseous Propellants / 266

7.7. Safety and Environmental Concerns / 267
Problems / 268

Symbols / 269
References / 269

8 Thrust Chambers 271

8.1. Injectors / 276

8.2. Combustion Chamber and Nozzle / 285

8.3. Low-Thrust Rocket Thrust Chambers or Thrusters / 301

8.4. Materials and Eabrication / 305

8.5. Heat Transfer Analysis / 311

8.6. Starting and Ignition / 323

8.7. Random Variable Thrust / 326

8.8. Sample Thrust Chamber Design Analysis / 328
Problems / 339

Symbols / 342
References / 344

9 Liquid Propellant Combustion and Its Stability 346

9.1. Combustion Process / 346

9.2. Analysis and Simulation / 350

9.3. Combustion Instability / 352
Problems / 364
References / 364

CONTENTS

10 Turbopumps and Their Gas Supplies 366

10.1. Introduction / 366

10.2. Descriptions of Several Turbopumps / 367

10.3. Selection of Turbopump Configuration / 372

10.4. Flow, Shaft Speeds, Power, and Pressure Balances / 376

10.5. Pumps / 378

10.6. Turbines / 387

10.7. Approach to Turbopump Preliminary Design / 391

10.8. Gas Generators and Preburners / 394
Problems / 395

Symbols / 396
References / 397

1 1 Engine Systems, Controls, and Integration 399

11.1. Propellant Budget / 399

11.2. Performance of Complete or Multiple Rocket Propulsion

Systems / 401

11.3. Engine Design / 403

11.4. Engine Controls / 411

11.5. Engine System Calibration / 423

11.6. System Integration and Engine Optimization / 430
Problems / 432
Symbols / 432
References / 433

Propellant Rocket Fundamentals 435

Basic Relations and Propellant Burning Rate / 437
Other Performance Issues / 458
Propellant Grain and Grain Configuration / 463
Propellant Grain Stress and Strain / 472

Attitude Control and Side Maneuvers with Solid Propellant Rocket
Motors / 483
Problems / 485
Symbols / 488
References / 489

13 Solid Propellants 492

13.1. Classihcation / 492

13.2. Propellant Characteristics / 498

12 Solid

12 . 1 .

12 . 2 .

12.3.

12.4.

12.5.

CONTENTS

13.3. Hazards / 505

13.4. Propellant Ingredients / 512

13.5. Other Propellant Categories / 522

13.6. Liners, Insulators, and Inhibitors / 526

13.7. Propellant Processing and Manufacture / 529
Problems / 532

References / 535

14 Solid Propellant Combustion and Its Stability 537

14.1. Physical and Chemical Processes / 537

14.2. Ignition Process / 541

14.3. Extinction or Thrust Termination / 543

14.4. Combustion Instability / 545
Problems / 553
References / 554

15 Solid Rocket Components and Motor Design 556

15.1. Motor Case / 556

15.2. Nozzles / 564

15.3. Igniter Hardware / 578

15.4. Rocket Motor Design Approach / 583
Problems / 590

References / 592

16 Hybrid Propellant Rockets 594

16.1. Applications and Propellants / 596

16.2. Interior Hybrid Motor Ballistics / 600

16.3. Performance Analysis and Grain Configuration / 604

16.4. Design Example / 609

16.5. Combustion Instability / 613
Problems / 617

Symbols / 618
References / 620

17 Electric Propulsion 622

17.1. Ideal Plight Performance / 628

17.2. Electrothermal Thrusters / 633

17.3. Nonthermal Electrical Thrusters / 640

17.4. Optimum Flight Performance / 656

CONTENTS

17.5. Mission Applications / 661

17.6. Electric Space-Power Supplies and Power-Conditioning

Systems / 663
Problems / 667
Symbols / 669
References / 671

18 Thrust Vector Control

18.1. TVC Mechanisms with a Single Nozzle / 675

18.2. TVC with Multiple Thrust Chambers or Nozzles / 685

18.3. Testing / 687

18.4. Integration with Vehicle / 687
Problems / 688
References / 689

19 Selection of Rocket Propulsion Systems

19.1. Selection Process / 693

19.2. Criteria for Selection / 699

19.3. Interfaces / 705
References / 709

20 Rocket Exhaust Plumes

20.1. Plume Appearance and Flow Behavior / 712

20.2. Plume Effects / 725

20.3. Analysis and Mathematical Simulation / 729
Problems / 730

References / 731

21 Rocket Testing

21.1. Types of Tests / 733

21.2. Test Facilities and Safeguards / 735

21.3. Instrumentation and Data Management / 742

21.4. Flight Testing / 746

21.5. Postaccident Procedures / 747
References / 748

673

691

710

733

CONTENTS

Appendix 1 Conversion Factors and Constants 749

Appendix 2 Properties of the Earth’s Standard Atmosphere 752

Appendix 3 Summary of Key Equations for Ideal Chemical

Rockets 753

Index 755

PREFACE

This eighth edition follows earlier editions and focuses on the subject of rocket
propulsion, its fundamentals, its essential technologies, and its key design ratio-
nale. It describes the various types of rocket propulsion systems, physical phe-
nomena, and essential relationships. It aims to balance a rigorous introduction of
fundamentals with tables, figures, and recent information that can be of interest
to practitioners. It endeavors to provide a clearer explanation of the basic physics
and chemistry of rocket propulsion and some of its design approaches.

Several chapters are devoted exclusively to liquid propellant rocket engines
and also to solid propellant rocket motors (almost half the book), the two most
prevalent of the rocket propulsion systems. It also has one chapter each on hybrid
propulsion (with one liquid and one solid propellant) and on electrical propul-
sion (with several different types). The principal flight applications for these
four major types of rocket propulsion systems are described. Brief discussions
of nuclear thermal propulsion systems and solar thermal propulsion system are
included. Our book does not contain speculative future propulsion schemes such
as nuclear fusion propulsion or pulsed detonations of propellant charges, because
these are as yet not considered to be practical.

The first few chapters of the book concern basic subjects. Following an intro-
duction are definitions of the key propulsion parameters, principal thermodynamic
relations, nozzle theory, vehicle flight analyses (flying inside and outside the
earth’s atmosphere), or thermochemical analyses of combustion gas properties.
The last few chapters relate intimately to rocket propulsion systems, namely thrust
vector control, selection of propulsion systems for specific missions, exhaust
plumes, and the testing of rocket propulsion systems.

The book has been used for six decades in undergraduate and graduate courses
of rocket propulsion in many universities, both in the United States and in other

xiii

XIV

PREFACE

countries. In this eighth edition many examples and problems have been revised
or added for a better discussion of the principles involved and the technical issues
used in preliminary design. More than 20 new homework problems have been
added (and several others were deleted) for more relevance of the text material
and for more practice of the key equations. Some of the problems have printed
answers, so students can self-check their solutions. Many of the problems are
patterned after the analyses and engineering efforts of actual propulsion systems.
A solutions manual is available to selected users. Most of the book may be
covered in a one- or two-semester course depending on the students’ background.

Since the first edition in 1949 it has been a most popular technical rocket
propulsion book and has been used by at least 75,000 students and professionals
in approximately 37 countries. According to a committee of the American Insti-
tute of Aeronautics and Astronautics it is the longest living aerospace book ever,
in print continuously for 60 years. It is the only aerospace book cited in two
prestigious book awards of the American Institute of Aeronautics and Astronau-
tics. Earlier editions have been translated into Russian, Chinese, and Japanese.
The authors have given guest lectures on rocket propulsion subjects at several
universities and a three-day course based on this book.

A number of new items and a few extensively changed sections appear in this
new eighth edition. For example, this is the first edition with a separate chapter
on turbopumps, an essential component of medium and large liquid propellant
rocket engines. It has new or extensively revised discussions on tank pressur-
ization, a new class of solid propellants that have been put into production, the
first high-thrust large hybrid propulsion system that has flown in an experimental
manned craft with nitrous oxide as the oxidizer and a rubbery fuel, a concept
for integrated computer programs for modern rocket propulsion engines, health
monitoring controls able to save hardware, and new flight applications for elec-
trical propulsion systems. Almost all libraries of rocket propulsion organizations
seem to have at least two editions of this book. The new edition has about three
dozen new and different figures, a dozen revised or new tables, and many other
clarifications, deletions, changes, or minor additions.

It is usually easier, faster, and less expensive to uprate or modify an existing
proven propulsion system for a new application than to design and build a truly
new one. For example, in the liquid propellant engine field with turbopump feed
systems, it has been more common in the last few decades to increase the thrust or
performance by modifying an existing engine than to undertake the development
of a brand new engine. A revised discussion of engine families (with related
engine designs) and a new table of 13 different models of the RF 10 upper-stage
engine emphasized this point.

Although the United States is officially committed to convert from the English
Engineering (EE) system of units (foot, pound) to the SI or metric system of units
(m, kg), this has not as yet happened. The major portion of the U.S. propulsion
industry for chemical rockets has been and is currently still using EE units for
design and manufacturing. If a particular customer wants SI units, they will be
given engineering information in SI units, but the fabrication will most likely

PREFACE

Still be done in EE units. Many colleges and research organizations in the United
States and most propulsion organizations in the other countries use the SI system
of units. So this book uses both systems; the examples and problems use one or
the other and some tables have both sets of units.

Computers have changed the way we do business in many fields. In rocket
propulsion computer programs have been developed for many engineering anal-
yses, computer-aided designs, computer-aided manufacturing, business and engi-
neering transactions, test data collection, data analysis, data presentation, project
management, and many others. In fact computers are used extensively in some
companies to design new propulsion systems. Therefore we identify in this book
the places where computer programs will be helpful and we mentions this often.
However, we do not discuss specific computer programs because they take up too
much space, become obsolete in a short time without regular upgrading, some
do not have a way to provide help to a user, and some of the better programs
are company proprietary and thus not publicly available.

Propellants are hazardous materials. The authors and the publisher recom-
mend that readers of this book do not work with them or handle them without
exhaustive study of the hazards, the behavior, and properties of each propellant,
and rigorous safety training including becoming familiar with protective equip-
ment. Safety training and propellant information is given routinely to employees
of organizations in this business. Neither the authors nor the publisher assume
any responsibility for actions on rocket propulsion taken by the readers, either
directly or indirectly. The information presented in this book is insufficient and
inadequate for conducting rocket propulsion experiments or operations.

The authors acknowledge gratefully the data, suggestions, and assistance given
to them while preparing the manuscript for this book. Fortunately we have had
the help from a number of outstanding experts in the field, reviewers, and edi-
tors. We appreciate the reviews of solid propellant motors by David McGrath
of ATK Launch Systems, Inc., and reviews of several chapters and the supply
of certain propellant data by Eckart W. Schmidt, formerly of Olin Aerospace
and currently a consultant in hazardous materials, is appreciated. Information on
electrical propulsion was obtained from Scott Benson, of NASA Glen Research
Center. Christopher Brophy of the Naval Post-Graduate School in Monterey made
suggestions on hybrid propulsion. Thanks to Thomas L. Moore formerly with
CPIAC (Chemical Propulsion Information Analysis Center) and now with ATK
Systems for a variety of information, such as the production of a new category of
solid propellant. Thanks also to Nick Heim of CPIAC for data on various motors,
permissible exposure limits for toxic vapors, and gelled propellants. Frederick S.
Simmons of The Aerospace Corporation reviewed the chapter on exhaust plumes
and provided additional data on plumes. James H. Morehart of The Aerospace
Corporation furnished precise detailed data on several rocket propulsion systems.
Aerojet General Corporation of Sacramento (they elected to remain anonymous
and did not single out specific deserving employees) provided data on a large
solid rocket booster. Jeffrey S. Kincaid of Pratt & Whitney Rocketdyne (PWR)
of Canoga Park, California, supplied the concept of integrated computer systems

XVI

PREFACE

and corrections to engine data, and Dennis Ramine of PWR in Canoga Park sent
information on heat transfer. Vince J. Wheelock formerly of Rocketdyne fur-
nished several key figures. Patricia Mills of PWR at West Palm Beach, Florida,
collected and sent data on the RL 10 engine family. Dan Magers of John Wiley
and Sons, Inc. updated references and the editors and printers of our publisher
made a coarse manuscript into a useful book.

George P. Sutton
Los Angeles, California

Oscar Biblarz,

Monterey, California

CHAPTER 1

CLASSIFICATION

The word propulsion comes from the Latin propulsus, which is the past participle
of the verb propellere, meaning to drive away. In a broad sense propulsion is the
act of changing the motion of a body. Propulsion mechanisms provide a force
that moves bodies that are initially at rest, changes a velocity, or overcomes
retarding forces when a body is propelled through a medium. Jet propulsion is
a means of locomotion whereby a reaction force is imparted to a device by the
momentum of ejected matter.

Rocket propulsion is a class of jet propulsion that produces thrust by ejecting
matter stored in a flying vehicle called the propellant. Duct propulsion is a class
of jet propulsion and includes turbojets and ramjets; these engines are also com-
monly called air-breathing engines. Duct propulsion devices utilize mostly the
surrounding medium as the “working fluid,” together with some vehicle-stored
fuel. Combinations of rockets and duct propulsion devices can be attractive for
some narrow applications and one is briefly described in this chapter.

The energy source most useful to rocket propulsion is chemical combustion.
Energy can also be supplied by solar radiation and, in the past, also by nuclear
reaction. Accordingly, the various propulsion devices can be divided into chem-
ical propulsion, nuclear propulsion, and solar propulsion. Table 1-1 lists many
of the important propulsion concepts according to their energy source and type
of propellant or working fluid. Radiation energy can originate from sources other
than the sun and theoretically can cover the transmission of energy by microwave
and laser beams, electromagnetic waves, and electrons, protons, and other par-
ticle beams from a transmitter to a flying receiver. Nuclear energy is associated
with the transformations of atomic particles within the nucleus of atoms and
usually is created by fission or fusion. Other energy sources, both internal (in the
vehicle) and external, can be considered. The energy form found in the output

2 CLASSIFICATION

TABLE 1-1. Energy Sources and Propellants for Various Propulsion Concepts

Propulsion Device

Energy Source"
Chemical Nuclear

Solar

Propellant or Working Fluid

Turbojet

D/P

TFD

Fuel -F air

Turbo-ramjet

TFD

Fuel -F air

Ramjet (hydrocarbon fuel)

D/P

TFD

Fuel -F air

Ramjet (H2 cooled)

TFD

Hydrogen -F air

Rocket (chemical)

D/P

TFD

Stored propellant

Ducted rocket

TFD

Stored solid fuel -F

surrounding air

Electric rocket

D/P

TFD

D/P

Stored propellant

Nuclear fission rocket

TFD

Stored H2

Nuclear fusion rocket

TFND

Stored H2

Solar-heated rocket

TFD

Stored H2

Photon rocket (hig light

TFND

Photon ejection (no stored

bulb)

propellant)

Solar sail

TFD

Photon reflection (no stored

propellant)

“D/P, developed and/or considered practical; TFD, technical feasibility has been demonstrated, but
development is incomplete; TFND, technical feasibility has not yet been demonstrated.

of a rocket is largely the kinetic energy of the ejected matter; thus the rocket
converts the input from the energy source into this form. The ejected mass can
be in a solid, liquid, or gaseous state. Often a combination of two or more of
these is ejected. At very high temperatures it can also be a plasma, which is an
electrically activated gas.

1.1. DUCT JET PROPULSION

This class, also called air-breathing engines, comprises devices which have a duct
to confine the flow of air. They use oxygen from the air to burn fuel stored in
the flight vehicle. The class includes turbojets, turbofans, ramjets, and pulsejets.
This class of propulsion is mentioned primarily to provide a comparison with
rocket propulsion and a background for combination rocket-duct engines, which
are mentioned later. Table 1-2 compares several performance characteristics
of specific chemical rockets with those of typical turbojets and ramjets. A high
specific impulse, which is a measure of performance to be defined later, is directly
related to a long flight range and thus indicates the superior range capability of air
breather engines over chemical rockets at relatively low altitude. The uniqueness
of the rocket, for example, high thrust to weight, high thrust to frontal area, and
nearly thrust independence of altitude, enables extremely long flight ranges to be
obtained in rarefied air and in space.

The turbojet engine is the most common of ducted engines. Figure 1-1 shows
the basic elements.

1.1. DUCT JET PROPULSION 3

TABLE 1-2. Comparison of Several Characteristics of a Typical Chemical Rocket and
Two-Duct Propulsion Systems

Feature

Chemical
Rocket Engine
or Rocket
Motor

Turbojet Engine

Ramjet Engine

Thrust-to-weight ratio,

75:1

5:1, turbojet and

7:1 at Mach 3 at

typical

afterburner

30,000 ft

Specific fuel consumption

8-14

0.5-1.5

2.3-3.5

(pounds of propellant or
fuel per hour per pound
of thrust)®

Specific thrust (pounds of

5000-25,000

2500 (low Mach at

2700 (Mach 2 at sea

thrust per square foot

sea level)

level)

frontal area)'’

Thrust change with altitude

Slight increase

Decreases

Thrust vs. flight speed

Nearly constant

Increases with

Increases with speed

Thrust vs. air temperature

Constant

speed

Decreases with

Flight speed vs. exhaust

Unrelated,

temperature
Flight speed

temperature
Flight speed always

velocity

flight speed

always less than

less than exhaust

can be

exhaust velocity

velocity

Altitude limitation

greater

None; suited to

14,000-17,000 m

20,000 m at Mach 3

Specific impulse, typical®

space travel
270 sec

1600 sec

30.000 m at Mach 5

45.000 m at Mach 12
1400 sec

(thrust force per unit
propellant or fuel weight
flow per second)

“Multiply by 0.102 to convert to kg/(hr-N).

'’Multiply by 47.9 to convert to N/m^

‘^Specific impulse is a performance parameter and is defined in Chapter 2.

At supersonic flight speeds above Mach 2, the ramjet engine (a pure duct
engine) becomes attractive for flight within the atmosphere. Thrust is produced
by increasing the momentum of the air as it passes through the ramjet, basically as
is accomplished in the turbojet and turbofan engines but without compressors or
turbines. Figure 1-2 shows the basic components of one type of ramjet. Ramjets
with subsonic combustion and hydrocarbon fuel have an upper speed limit of
approximately Mach 5; hydrogen fuel, with hydrogen cooling, raises this to at
least Mach 16. Ramjets with supersonic combustion are known as scramjets and
have flown in experimental vehicles. All ramjets depend on rocket boosters or
some other method (such as being launched from an aircraft) for being accelerated

4 CLASSIFICATION

FIGURE 1-1. Simplified schematic diagram of a turbojet engine.

Fuel

FIGURE 1-2. Simplified diagram of a ramjet with a supersonic inlet (converging and
diverging flow passage).

to near their design flight speed to become functional. The primary applications of
ramjets with subsonic combustion have been in shipboard and ground-launched
antiaircraft missiles. Studies of a hydrogen-fueled ramjet for hypersonic aircraft
look promising. The supersonic flight vehicle is a combination of a ramjet-driven
high-speed airplane and a one- or two-stage rocket booster. It can travel at speeds
up to a Mach number of 25 at altitudes of up to 50,000 m.

1.2. ROCKET PROPULSION

Rocket propulsion systems can be classified according to the type of energy
source (chemical, nuclear, or solar), the basic function (booster stage, sustainer
or upper stages, attitude control, orbit station keeping, etc.), the type of vehicle
(aircraft, missile, assisted takeoff, space vehicle, etc.), size, type of propellant,
type of construction, or number of rocket propulsion units used in a given vehicle.

Another way is to classify by the method of producing thrust. A thermody-
namic expansion of a gas is used in the majority of practical rocket propulsion
concepts. The internal energy of the gas is converted into the kinetic energy of
the exhaust flow and the thrust is produced by the gas pressure on the surfaces
exposed to the gas, as will be explained later. This same thermodynamic the-
ory and the same generic equipment (nozzle) is used for jet propulsion, rocket
propulsion, nuclear propulsion, laser propulsion, solar-thermal propulsion, and
some types of electrical propulsion. Totally different methods of producing thrust

1.2. ROCKET PROPULSION 5

are used in other types of electric propulsion or by using a pendulum in a gravity
gradient. As described below, these electric systems use magnetic and/or electric
fields to accelerate electrically charged molecules or atoms at very low densities.
It is also possible to obtain a very small acceleration by taking advantage of the
difference in gravitational attraction as a function of altitude, but this method is
not explained in this book.

The Chinese developed and used solid propellant in rocket missiles over 800
years ago, and military bombardment rockets were used frequently in the eigh-
teenth and nineteenth centuries. However, the significant developments of rocket
propulsion took place in the twentieth century. Early pioneers included the Rus-
sian Konstantin E. Ziolkowsky, who is credited with the fundamental rocket
flight equation and his 1903 proposals to build rocket vehicles. The German
Hermann Oberth developed a more detailed mathematical theory; he proposed
multistage vehicles for space flight and fuel-cooled thrust chambers. The Amer-
ican Robert H. Goddard is credited with the first flight using a liquid propellant
rocket engine in 1926. An early book on the subject was written by the Viennese
engineer Eugen Sanger. For rocket history see Refs. 1-1 to 1-8.

Chemical Rocket Propulsion

The energy from a high-pressure combustion reaction of propellant chemicals,
usually a fuel and an oxidizing chemical, permits the heating of reaction product
gases to very high temperatures (2500 to 4100°C or 4500 to 7400°F). These
gases subsequently are expanded in a nozzle and accelerated to high velocities
(1800 to 4300 m/sec or 5900 to 14,100 ft/sec). Since these gas temperatures are
about twice the melting point of steel, it is necessary to cool or insulate all the
surfaces that are exposed to the hot gases. According to the physical state of
the propellant, there are several different classes of chemical rocket propulsion
devices.

Liquid propellant rocket engines use liquid propellants that are fed under pres-
sure from tanks into a thrust chamber.* A typical pressure-fed liquid propellant
rocket engine system is schematically shown in Fig. 1-3. The liquid bipropellant
consists of a liquid oxidizer (e.g., liquid oxygen) and a liquid fuel (e.g., kerosene).
A monopropellant is a single liquid that contains both oxidizing and fuel species;
it decomposes into hot gas when properly catalyzed. A large turbopump-fed liq-
uid propellant rocket engine is shown in Fig. 1-4. Gas pressure feed systems are
used mostly on low thrust, low total energy propulsion systems, such as those
used for attitude control of flying vehicles, often with more than one thrust cham-
ber per engine. Pump-fed liquid rocket systems are used typically in applications
with larger amounts of propellants and higher thrusts, such as in space launch
vehicles. See Refs. 1-1 to 1-6.

*The term thrust chamber, used for the assembly of the injector, nozzle, and chamber, is preferred
by several official agencies and therefore has been used in this book. For small spacecraft control
rockets the term thruster (a small thrust chamber) is commonly used, and this term will be used in
some sections of this book.

6 CLASSIFICATION

Check Check

valve valve

FIGURE 1-3. Schematic flow diagram of a liquid propellant rocket engine with a gas
pressure feed system. The dashed lines show a second thrust chamber, but some engines
have more than a dozen thrust chambers supplied by the same feed system. Also shown
are components needed for start and stop, controlling tank pressure, filling propellants
and pressurizing gas, draining or flushing out remaining propellants, tank pressure relief
or venting, and several sensors.

In the thrust chamber the propellants react to form hot gases, which in turn are
accelerated and ejected at a high velocity through a supersonic nozzle, thereby
imparting momentum to the vehicle. A nozzle has a converging section, a con-
striction or throat, and a conical or bell-shaped diverging section as further
described in the next two chapters.

Some liquid rocket engines permit repetitive operation and can be started and
shut off at will. If the thrust chamber is provided with adequate cooling capacity,
it is possible to run liquid rockets for periods exceeding 1 hour, dependent only on

1.2. ROCKET PROPULSION 7

FIGURE 1-4. Simplified schematic diagram of one type of liquid propellant rocket
engine with a turbopump feed system and a separate gas generator, which generates
“warm” gas for driving the turbine. Not shown are components necessary for controlling
the operation, filling, venting, draining, or flushing out propellants, filters or sensors. This
turbopump assembly consists of two propellant pumps, a gear case, and a high speed
turbine.

the propellant supply. A liquid rocket propulsion system requires several precision
valves and a complex feed mechanism which includes propellant pumps, turbines,
or a propellant-pressurizing device, and a relatively intricate combustion or thrust
chamber.

In solid propellant rocket motors* the propellant to be burned is contained
within the combustion chamber or case (see Fig. 1-5). The solid propellant

‘Historically, the word engine is used for a liquid propellant rocket propulsion system and the word
motor is used for solid propellant rocket propulsion. They were developed originally by different
groups.

8 CLASSIFICATION

Nozzle throat

insert Nozzle exit cone

FIGURE 1-5. Simplified perspective three-quarter section of a typical solid propellant
rocket motor with the propellant grain bonded to the case and the insulation layer and
with a conical exhaust nozzle. The cylindrical case with its forward and aft hemispherical
domes form a pressure vessel to contain the combustion chamber pressure. Adapted with
permission from Ref. 12-1.

charge is called the grain and it contains all the chemical elements for complete
burning. Once ignited, it usually bums smoothly at a predetermined rate on all the
exposed internal surfaces of the grain. Initial burning takes place at the internal
surfaces of the cylinder perforation and the four slots. The internal cavity grows
as propellant is burned and consumed. The resulting hot gas flows through the
supersonic nozzle to impart thrust. Once ignited, the motor combustion proceeds
in an orderly manner until essentially all the propellant has been consumed. There
are no feed systems or valves. See Refs. 1-7 to 1-10.

Liquid and solid propellants, and the propulsion systems that use them, are dis-
cussed in Chapters 6 to 11 and 12 to 15, respectively. Liquid and solid propellant
rocket propulsion systems are compared in Chapter 19.

Gaseous propellant rocket engines use a stored high-pressure gas, such as air,
nitrogen, or helium, as their working fluid or propellant. The stored gas requires
relatively heavy tanks. These cold gas engines have been used on many early
space vehicles for low thmst maneuvers and for attitude control systems and
some are still used today. Heating the gas by electrical energy or by combus-
tion of certain monopropellants improves the performance and this has often
been called warm gas propellant rocket propulsion . Chapter 7 discusses gaseous
propellants.

Hybrid propellant rocket propulsion systems use both a liquid and a solid
propellant. For example, if a liquid oxidizing agent is injected into a combus-
tion chamber filled with a solid carbonaceous fuel grain, the chemical reaction
produces hot combustion gases (see Fig. 1-6). They are described further in
Chapter 16. Several have flown successfully.

1.2. ROCKET PROPULSION 9

FIGURE 1-6. Simplified schematic diagram of a typical hybrid rocket engine. The
relative positions of the oxidizer tank, high-pressure gas tank, and the fuel chamber with
its nozzle depend on the particular vehicle design.

Combinations of Ducted Jet Engines and Rocket Engines

The Tomahawk surface-to-surface missile uses two stages of propulsion in
sequence. The solid propellant rocket booster lifts the missile away from its
launch platform and is discarded after its operation. A small turbojet engine
sustains the low-level flight at nearly constant speed toward the target.

A ducted rocket, sometimes called an air-augmented rocket, combines the
principles of rocket and ramjet engines; it gives higher performance (specific
impulse) than a chemical rocket engine, while operating within the earth’s atmo-
sphere. Usually the term air-augmented rocket denotes mixing of air with the
rocket exhaust (fuel rich for afterburning) in proportions that enable the propul-
sion device to retain the characteristics typifying a rocket engine, for example,
high static thrust and high thrust-to-weight ratio. In contrast, the ducted rocket
often is like a ramjet in that it must be boosted to operating speed and uses the
rocket components more as a fuel-rich gas generator (liquid or solid).

The principles of the rocket and ramjet can be combined. An example of
these two propulsion systems operating in sequence and in tandem and yet
utilize a common combustion chamber volume as shown in Fig. 1-7. The low-
volume configuration, known as an integral rocket-ramjet , can be attractive in
air-launched missiles using ramjet propulsion. The transition from the rocket
to the ramjet requires enlarging the exhaust nozzle throat (usually by ejecting
rocket nozzle parts), opening the ramjet air inlet-combustion chamber interface,
and following these two events with the normal ramjet starting sequence.

A solid fuel ramjet uses a grain of solid fuel that gasifies or ablates and
reacts with air. Good combustion efficiencies have been achieved with a patented
boron-containing solid fuel fabricated into a grain similar to a solid propellant
and burning in a manner similar to a hybrid rocket propulsion system.

10 CLASSIFICATION

Solid rocket

Fuel manifold-. propellant \

Ejectable

^ rocket
^nozzle
inserts

^ ^Ramjet
nozzle

FIGURE 1-7. Elements of an air-launched missile with integral rocket-ramjet propul-
sion. After the solid propellant has been consumed in boosting the vehicle to flight speed,
the rocket combustion chamber becomes the ramjet combustion chamber with air burning
the ramjet liquid fuel.

Nuclear Rocket Engines

Two different types of nuclear energy sources have been investigated for deliv-
ering heat to a working fluid, usually liquid hydrogen, which subsequently can
be expanded in a nozzle and thus accelerated to high ejection velocities (6000
to 11,000 m/sec). However, none can be considered fully developed today and
none have flown. They are ihe, fission reactor and the fusion reactor. Both are
basically extensions of liquid propellant rocket engines. The heating of the gas is
accomplished by energy derived from transformations within the nuclei of atoms.
In chemical rockets the energy is obtained from within the propellants, but in
nuclear rockets the power source is usually separate from the propellant.

In the nuclear fission reactor rocket, heat can be generated by the fission of
uranium in the solid reactor material and subsequently transferred to the work-
ing fluid (see Refs. 1-11 to 1-13). The nuclear fission rocket is primarily a
high- thrust engine (above 40,000 N) with specific impulse values up to 900 sec.
Fission rockets were designed and tested in the 1960s in the United States and
also in the Soviet Union, which today is Russia. Ground tests with hydrogen
as a working fluid culminated in a thrust of 980,000 N (210,0001b force) at a
graphite core nuclear reactor power level of 4100 MW with an equivalent altitude-
specific impulse of 848 sec and a hydrogen temperature of about 2500 K. There
were concerns with the endurance of the materials at the high temperature (above
2600 K) and intense radiations, power level control, cooling a reactor after oper-
ation, moderating the high-energy neutrons, and designing lightweight radiation
shields for a manned space vehicle. No further ground tests of nuclear fission
rocket engines have been undertaken.

In recent years there has been renewed interest in nuclear fission rocket propul-
sion primarily for a potential manned planetary exploration mission. Studies
have shown that the high specific impulse (estimated in some studies at 1100
sec) allows shorter interplanetary trip transfer times, smaller vehicles, and more
flexibility in the launch time when planets are not in their optimum relative
position. See Refs. 1-11 and 1-13.

1.2. ROCKET PROPULSION

Fusion is an alternate way to create nuclear energy, which can heat a working
fluid. A number of different concepts have been studied. To date none are feasible
or practical.

Concerns about an accident with the inadvertent spreading of radioactive mate-
rials in the earth environment and the high cost of development programs have
to date prevented a renewed experimental development of a large nuclear rocket
engine. Unless there are some new findings and a change in world attitude about
nuclear radiation, it is unlikely that a nuclear rocket engine will be developed or
flown in the next few decades. Therefore no further discussion of it is given in
this book.

Electric Rocket Propulsion

In all electric propulsion the source of the electric power (nuclear, solar radiation
receivers, or batteries) is physically separate from the mechanism that produces
the thrust. This type of propulsion requires a heavy and inefficient power sources
as discussed below. The thrust usually is low, typically 0.005 to 1 N. In order to
allow a significant increase in the vehicle velocity, it is necessary to apply the
low thrust and thus a small acceleration for a long time (weeks or months) (see
Chapter 17 and Refs. 1-14 and 1-15).

Of the three basic types, electrothermal rocket propulsion most resembles
the previously mentioned chemical rocket units; propellant is heated electrically
(by heated resistors or electric arcs), and the hot gas is then thermodynamically
expanded and accelerated to supersonic velocity through an exhaust nozzle (see
Fig. 1-8). These electrothermal units typically have thrust ranges of 0.01 to 0.5 N,
with exhaust velocities of 1000 to 5000 m/sec, and ammonium, hydrogen, nitro-
gen, or hydrazine decomposition product gases have been used as propellants.

The two other types — the electrostatic or ion propulsion engine and the
electromagnetic or magnetoplasma engine — accomplish propulsion by differ-
ent principles, and the thermodynamic expansion of gas in a nozzle, as such,
does not apply. Both will work only in a vacuum. In an ion rocket (see Fig. 1-9)
a working fluid (typically, xenon) is ionized (by stripping off electrons), and then
the electrically charged heavy ions are accelerated to very high velocities (2000
to 60,000 m/sec) by means of electrostatic fields. The ions are subsequently elec-
trically neutralized; they are combined with electrons to prevent the buildup of
a space charge on the vehicle.

In the magnetoplasma rocket an electrical plasma (an energized hot gas con-
taining ions, electrons, and neutral particles) is accelerated by the interaction
between electric currents and magnetic fields and ejected at high velocity (1000
to 50,000 m/sec). There are many different types and geometries. The Hall-effect
thruster, which accelerates a plasma, has a good flight record in Russia. A simple
pulsed (not continuously operating) electrical propulsion unit with a solid pro-
pellant is shown in Fig. 1-10. It has had a good flight record as a spacecraft
attitude control engine.

12 CLASSIFICATION

Arc between cathode
tip and annular
region of anode

FIGURE 1-8. Simplified schematic diagram of arc-heating electric rocket propulsion
system. The arc plasma temperature is very high (perhaps 15,000 K) and the anode,
cathode, and chamber will get hot (1000 K) due to heat transfer.

FIGURE 1-9. Simplified schematic diagram of a typical ion rocket, showing the approx-
imate distribution of the electric power.

Other Rocket Propulsion Concepts

Several teehnologies exist for harnessing solar energy to provide the power for
spacecraft and also to propel spacecraft using electrical propulsion. Solar cells
generate electric power from the sun’s radiation. They are well developed and
have been successful for several decades. Most electric propulsion systems have
used solar cells for their power supply. Batteries and isotope decay power sources
have also been used.

One concept is the solar thermal racket', it has large-diameter optics to concen-
trate the sun’s radiation (e.g., by lightweight precise parabolic mirrors or Fresnel
lenses) onto a receiver or optical cavity. Figure 1-11 shows one concept and
some data is given in Table 2-1. The receiver is made of high-temperature metal
(such as tungsten or rhenium) and has a cooling jacket or heat exchanger. It
heats a working fluid, usually liquid hydrogen, up to perhaps 2500°C and the hot
gas is controlled by hot gas valves and exhausted through one or more nozzles.
The large mirror has to be pointed toward the sun, and this usually requires the
mirror to be adjustable in its orientation as the spacecraft orbits around the earth.

1.2. ROCKET PROPULSION 13

Plasma exhaust from
parallel rail nozzle

FIGURE 1-10. Simplified diagram of a rail accelerator for self-induced magnetic accel-
eration of a current-carrying plasma. When the capacitor is discharged, an arc is struck at
the left side of the rails. The high current in the plasma arc induces a magnetic field. The
action of the current and the magnetic field causes the plasma to be accelerated at right
angles to both the magnetic held and the current, namely in the direction of the rails.
Each time the arc is created a small amount of solid propellant (Teflon) is vaporized and
converted to a small plasma cloud, which (when ejected) gives a small pulse of thrust.
Actual units can operate with many pulses per second.

Heat receiver

Radiation
from the
sun

FIGURE 1-11. Simplified schematic diagram of a solar thermal rocket concept.

Performance can be two to three times higher than that of a chemical rocket and
thrust levels in most studies are low (1 to 10 N). Since large lightweight optical
elements cannot withstand drag forces without deformation, the optical systems
are deployed outside the atmosphere. Contamination is negigible, but storage or
refueling of liquid hydrogen is a challenge. Problems being investigated include

14 CLASSIFICATION

rigid, lightweight mirror or lens structures, operational life, minimizing hydrogen
evaporation, and heat losses to other spacecraft components. To date the solar
thermal rocket has not yet provided the principal thrust of a flying spacecraft.

The solar sail is another concept. It is basically a big photon reflector surface.
The power source for the solar sail is the sun and it is external to the vehicle
(see Ref. 1-16). Approaches using nuclear explosions and pulsed nuclear fusion
have been analyzed (Refs. 1-17 and 1-18) but are not yet feasible. Concepts
for transmitting radiation energy (by lasers or microwaves) from ground stations
on Earth to satellites have been proposed but are not yet developed.

International Rocket Propulsion Effort

Active development or production of rocket propulsion systems was or is under
way in more than 30 different countries. Some of them have made signihcant
and original contributions to the state of the art of the technologies. There is
mention in this book of a few foreign rocket units and their accomplishments
and references to international rocket literature. Although most of the data in
this book are taken from U.S. rocket experience, this is not intended to minimize
foreign achievements.

At the time of this writing the major international program has been the
International Space Station (ISS), a multiyear cooperative effort with major con-
tributions from the United States and Russia and active participation by several
other nations. This manned orbital space station is used for conducting experi-
ments and observations on a number of research projects. See Ref. 1-19.

1.3. APPLICATIONS OF ROCKET PROPULSION

Because the rocket can reach a performance unequaled by other prime movers,
it has its own fields of application and does not usually compete with other
propulsion devices. Examples of important applications are given below and
discussed further in Chapter 4.

Space Launch Vehicles

Between the first space launch in 1957 and the end of 1998 approximately 4102
space launch attempts have taken place in the world and all but about 129 were
successful (see Ref. 1-20). Space launch vehicles or space boosters can be clas-
sified broadly as expendable or recoverable/reusable. Other bases of classification
are the type of propellant (storable or cryogenic liquid or solid propellants), num-
ber of stages (single-stage, two-stage, etc.), size/mass of payloads or vehicles,
and manned or unmanned.

Each space launch has a specific space flight objective, such as an earth orbit
or a moon landing. See Ref. 1-21. It uses between two and hve stages, each
with its own propulsion systems, and each is usually fired sequentially after the

1 .3. APPLICATIONS OF ROCKET PROPULSION 1 5

lower stage is expended. The number of stages depends on the specific space
trajectory, the number and types of maneuvers, the energy content of a unit mass
of the propellant, payload size, and other factors. The initial stage, usually called
the booster stage, is the largest and it is operated first; this stage is then separated
from the ascending vehicle before the second-stage rocket propulsion system is
ignited and operated. As will be explained in Chapter 4, adding an extra stage
permits a significant increase in the payload (such as more scientific instruments
or more communications gear).

Each stage of a multistage launch vehicle is essentially a complete vehicle in
itself and carries its own propellant, its own rocket propulsion system or systems,
and its own control system. Once the propellant of a given stage is expended, the
remaining mass of that stage (including empty tanks, cases, structure, instruments,
etc.) is no longer useful in providing additional kinetic energy to the succeeding
stages. By dropping off this useless mass it is possible to accelerate the final
stage with its useful payload to a higher terminal velocity than would be attained
if multiple staging were not used. Both solid propellant and liquid propellant
rocket propulsion systems have been used for low earth orbits.

Figure 1-12 shows the Delta IV heavy-lift space launch vehicle at takeoff.
Its propellants are liquid oxygen/liquid hydrogen (LOX/LH 2 ) in all its main
engines. Its booster engine, the Pratt & Whitney Rocketdyne RS-68, is shown
in Fig. 6-10 and data is in Table 11-2 and its second-stage engine, the Pratt
& Whitney Rocketdyne RL lOB-2 LOX/LH 2 (24,7501b thrust) is shown in Fig.
8-17 and data is in Table 8-1. The two liquid propellant strap-on booster pods
(with the same booster engine) are removed for smaller payloads. Figure 1-13
shows the Atlas V space launch vehicle. Its booster engine is the Energomash
RD-I80, it has Aerojet solid propellant strap-on boosters, and the upper stage
engine is the Pratt & Whitney Rocketdyne RL lOA-4-2 LOX/LH 2 engine. The
Russian (Energomash) LOX/kerosene RD-180 engine is shown in Ref. 1-2 as its
Figure 7.10-1 1 and data is in its Table 7.10-2. In both of these two launch vehicles
the payload is carried on top of the second stage and it has its own propulsion
system of small thrusters. Table 1-3 gives data of the larger propulsion systems
of these two U.S. launch vehicles.

The U.S. Space Shuttle is one of the older programs and it provided the
first reusable spacecraft that lands on a runway. Figure 1-14 shows the basic
configuration of the Space Shuttle, which consists of two stages, the booster and
the orbiter. It shows all the 67 rocket propulsion systems of the shuttle. The
orbiter is really a reusable combination space launch vehicle, spacecraft, and
glider for landing. The two solid propellant strap-on rocket motors are the largest
in existence; they are equipped with parachutes for sea recovery of the bumed-out
motors. The large LO 2 /LH 2 external tank is jettisoned and expended just before
orbit insertion (see Ref. 1-22). Details of several of these Space Shuttle rocket
propulsion systems are given in Table 1 -4. The Space Shuttle accomplishes both
civilian and military missions of placing satellites in orbit, undertaking scientific
exploration, supplying a space station, and repairing, servicing, and retrieving
satellites.

16 CLASSIFICATION

FIGURE 1-12. Heavy lift Delta IV space launch vehicle. The center liquid propellant
booster stage has a Pratt & Whitney Rocketdyne RS-68 rocket engine (LOX/LH 2 ). The
two strap-on stages each use the same engine. (Courtesy Pratt & Whitney Rocketdyne)

1.3. APPLICATIONS OF ROCKET PROPULSION

FIGURE 1-13. Atlas V space launch vehicle with three (or five) strap-on stages using
Aerojet solid rocket motors and a central Energomash (Russia) RD-180 liquid propellant
booster rocket engine running on LOX/kerosene. (Courtesy Lockheed Martin Corp.)

TABLE 1-3. Propulsion System Data for Two Major U.S. Lauch Vehicles

^ -2 s

(u 3

S <

o ti

Z ><

D CD

s s

^ 2

U (X

h —

.H 3 S

D ^ p_, o

3 3 o E

X O ^

O D

fj~ IZ3 ,

s a

^ £ '
c/5 h3

^ flj

.2 S M

tfl C cd

3 -2 c/3
S izi y

g -3 S

“ I ^

s 2

Q <=^

»/l

»o

VJ ■,'

cd >-J
> C/5

C/5

Cli

0^:

■a

!C 00

_ o

m ^

'sd

C CN

Z m

c/5

Cii

»o

»4

4.9

»o

'flj

c/3

'oo

ci a

::^

o S

O 3

o ^
8 5

D-

U-3 £

s 'a

'd-

t"-
m '

r-"

(SL)

’d

■4

'(SL) means sea level.

At ignition.

■RD-180 has 2 gimbal mounted thrust chambers.

FIGURE 1-14. Simplified sketch of original version of Space Shuttle vehicle. The shuttle orbiter — the delta-winged vehicle about the size of
a medium-range jet liner — is a reusable, cargo carrying, spacecraft-airplane combination that takes off vertically and lands horizontally like a
glider. Each shuttle orbiter was designed for a minimum of 100 missions and can carry as much as 65,0001b of payload to a low Earth orbit,
and a crew of up to four members and 10 passengers. It can return up to 25,0001b of payload back to earth.

TABLE 1-4. Propulsion Systems for the Space Shuttle

§ ^

.ii H

^ c:

z ^

o c =
Z <u

s- °

■Q cj

£)

<rs

6 -e

w) o

-S 3

00 T3

U 1)
3 —

.S 3

U ”3

IS o

C O

■3 -a

<D u (U £f'

O-

' x>

-£ c o

q:;' c u o 'co ^

r- 3 o =i

2 ^

-3 T3

■g ■=

'3 .S'

Z in

'tT

CO 3 in

S •§ ji

^ m

o .«>
O "O

o -a o ■£ "S

■3 s- ^

C _ o '3 O

w S o -s 3 p

>■ 52 is ^ ^ ^

— .3, = 'C ~o

5 -2 R c 2 S

o E 3

r- S

2£ > -7

lis

P o

•P

^ ^ ^
R 3 £

3 ^

°S ^

^ "S

3 CO
.3 (U

2 >n

o 3 :

<U N
cfl N

m c

^ CN
O

II R

I ■g § s o

ifc

!C 3 S -s ^

O £ .9- g 2 =

— a a>

>> H

« QS
rT '*-

■S £

O rn

CM CN

3 r4

a II
-o ^

I °

£ - -a
« 13 .S

— 3

XI .00

“ -S

ctS 2
00 p

1. MMH, monomethylhydrazine and NTO, nitiogen tetroxide.

2. 70% Ammonium perchlorate; 16% aluminum; 12% polybutadiene acrylic acid binder; 2% epoxy curing agent.

1 .3. APPLICATIONS OF ROCKET PROPULSION 21

Table 1-4 gives propulsion system data on the Space Shuttle, which is really
a combination of launch vehicle, spacecraft, and a glider. It can be seen that
the thrust levels are highest for booster or first stages and are relatively high for
upper stages (thousands of pounds). Only for the attitude control system of the
vehicle (also called reaction control in Table 1 -4) are the thrust levels low (from
a fraction of a pound for small spacecraft to as high as about 1000 pounds thrust
in the Space Shuttle vehicle). Frequent propulsion starts and stops are usually
required in these applications.

At the time of this writing (early 2008) for this new 8th edition National
Aeronautics and Space Administration (NASA) had awarded the initial contracts
for a new large manned space flight vehicle identified as Ares I. It is intended
to replace the aging Space Shuttle after about 2012. It is planned to use a large
single 5-segment solid rocket motor as booster propulsion (being developed by
ATK Launch Systems) and the second stage will use a J-2X liquid propellant
rocket engine with LOX/LH 2 propellants (being developed by Pratt & Whitney
Rocketdyne).

A new set of manned space flight vehicles are currently being developed by
several entrepreneurial US and foreign companies, many with their own private
capital. They are aimed at the future commercial market of sending tourists
(for a hefty price) into space and returning them safely back to earth. All are
based on reusable spacecraft, some with reusable launch vehicles, some with
vertical and some with horizontal takeoff or landing. A couple of suhorbital flights
have already been accomplished with professional pilots using a winged vehicle.
Liquid propellant rocket engines seem to be preferred and many are planned to
be reusable. It is too early for determining which of these organizations will be
successful in commercializing manned space flight.

The missions and payloads for space launch vehicles are many, such as
military (reconnaissance satellites, command and control satellites), nonmili-
tary government [weather observation satellites, global positioning system (GPS)
satellites], space exploration (space environment, planetary missions), or com-
mercial (communication satellites).

A single stage-to-orbit vehicle, attractive because it avoids the costs and com-
plexities of staging, has been expected to have improved reliability (simpler
structures, fewer components). However, its payload is usually too small. A low
earth orbit (say 100 miles altitude) can only be achieved with such a vehicle
if the propellant performance is very high and the structure is efficient and low
in mass. Liquid propellants such as liquid hydrogen with liquid oxygen were
usually chosen. To date a large rocket-propelled single stage-to-orbit vehicle has
not flown.

Spacecraft

Depending on their missions, spacecraft can be categorized as earth satellites,
lunar, interplanetary, and trans-solar types, and as manned and unmanned
spacecraft. Reference 1-23 lists over 20,000 satellites and categorizes them as
satellites for communications, weather, navigation, scientific exploration, deep

22 CLASSIFICATION

space probes, observation (including radar surveillance), reconnaissance and
other applications. Rocket propulsion is used for both primary propulsion (i.e.,
acceleration along the flight path, such as for orbit insertion or orbit change
maneuvers) and secondary propulsion functions in these vehicles. Some of the
secondary propulsion functions are attitude control, spin control, momentum
wheel and gyro unloading, stage separation, and the settling of liquids in tanks.
A spacecraft usually has a series of different rocket propulsion systems, some
often very small. For spacecraft attitude control about three perpendicular axes,
each in two rotational directions, the system must allow the application of pure
torque for six modes of angular freedom, thus requiring a minimum of 12
thrusters. Some missions require as few as 4 to 6 thrusters, whereas the more
complex manned spacecraft have 40 to 80 thrusters in all of its stages. Often
the small attitude control rockets must give pulses or short bursts of thrust,
necessitating thousands of restarts. See Section 6.7 and Ref. 1-24.

Table 1-5 presents a variety of spacecraft along with their weights, missions,
and propulsion. Although only U.S. launch vehicles are listed in this table, there
are also launch vehicles developed by France, the European Space Agency, Rus-
sia, Japan, China, India, and Israel that have successfully launched payloads into
satellite orbits. They use rocket propulsion systems that were developed in their
own countries.

The majority of spacecraft have used liquid propellant engines, a few with solid
propellant boosters. Many spacecraft have operated successfully with electrical
propulsion for attitude control. Electrical propulsion systems will probably also
be used for some primary and secondary spacecraft propulsion missions on long-
duration space flights, as described in Chapter 17.

Micropropulsion is a relatively new designation. It has been deflned as any
rocket propulsion system that is applicable to small spacecraft with a mass of
less than 100 kg, or 220 lb. See Ref. 1-25. It encompasses a variety of different
propulsion concepts, such as certain very low thrust liquid mono- and bipropellant
rocket engines, small gaseous propellant rocket engines, several types of electrical
propulsion systems, and emerging advanced versions of these. Many are based
on fabrication of very small components (valves, thrusters, switches, insulators,
or sensors) by micromachining and electromechanical processes.

Missiles and Other Applications

Military missiles can be classified as shown in Table 1-6. Rocket propulsion for
new U.S. missiles uses now almost exclusively solid propellant rocket motors.
They can be strategic missiles, such as long-range ballistic missiles (800 to
9000 km range), which are aimed at military targets within an enemy country, or
tactical missiles, which are intended to support or defend military ground forces,
aircraft, or navy ships.

Tables 1-6 and 1-7 show some parameters of rocket propulsion devices
for different applications. The selection of the best rocket propulsion system
type and design for any given application is a complex process involving many
factors, including system performance, reliability, propulsion system size, and
compatibility, as described in Chapter 19.

Space Maneuver Propulsion

-Q

■a c

Oh •'S
D X

too

•S

’B

(U 2

■y C

CO 3

■£ c

cO C
ja CO

2 M

u S

^ C

a .

M c

a o

CJ Dh

X o

O X

X D

-e'o^

T3 c/3
CO ID
0^: 0^

OJ)

c ^

X o

Q-

^ (D

D 2

X .2

g d .

•c “ ^

CO d ^

• o

c/3

c/i

^ a

U Q,

'+^ c

’3 E

Ci- o

.SP

0 0

X-

0 0

I> wo

wo"

c c d c c

cO cO cO cO cO

D D D D
Oh O- Oh D.

O O O O

O- O-
O O

Oh D-
O O

c c

O D
C (30

e o

o _

D D D D D X

_C _C _C

N N N N N .

cO cO cO cO cO X

l- ^ ^ 5- Vh D

-O »!■
>> 2
i 00

T3 -a

X E

-a X

X ffi

X cs

f 5 i ffi
s i i s §
^ ° Is s

o § ^O o

Z ^

<N (N

' z z

i-i

K K E

-d X

X E

b s

•c

D-

C/5

^ ^

c/3

S/J

!>.

C/i

b-s

L' 3 '3

c/3

-a

3 c=

g c C c

§ c 0 0 3?

2 ‘u 0 y _

eg)

'+-(

s ®

3 0

CHh

c/3

'C Dh <0

o 8 p

wo X wo

D O
^ O
WO

O (N

D CJ)

c/3

2 c

0 X

E >

C/5

wo ^
(N ^

00 ^

CO c/5 O

o ^
^ d

D CJ

C/5

D >

O _
O D

n , .H

3 (i>

c/5

‘'N2O4, nitrogen tetroxide (oxidizer); MMH, monomethylhydrazine (fuel); 50;50 UDMH-N2H4 is a 50% mixture of unsymmetrical dimethylhydrazine and
hydrazine.

TABLE 1-6. Selected Data for U.S. Missiles

00 o

»n

, ,

-H in

— H

1-^

Os^

00^

'd-

CM^

1-H

'ct

_oo

od'

cm"

•o

-o

■a

T 3

C /2

T 3

d 3

d 5

d 3

*02

c /2

c/2

C/f

c/2

C/ 2 "

c/2

C/i

c/2

■a

c/2

C/i

c/2

C/i

C/f

•2

-2

c/2

CO CM — — CN

00 in »o

OS ^ Os 'sd

m ^ ^

CM CM ^ CM
m:> SO d — '

O^^coOScMCMO
— ^ -H ^ CM CM

^ ^ O
CM m o

i> so in CM in

sO'^socM^sor^co

S "a

.u-

U .2
TO in
- W td
c/5 Cu

■d

^ $ o

• ^ < CJ
J= Cd

c/5 c/5 pin

■| g o

^ Oh D
c/5 c/5 c/5

0J5

(U
CJ

^ ,cd

Vh 4-(

O id

o I 3

a ‘f

■c S 2

I 2 8

»n CO

d cd

00
D c

S-H .S

CC ^

C/5

Tow (antitank) 0.58 3.84 1 stage, solid 40

Cruise missile (subsonic) Tomahawk 1.74 21 solid booster + 3,900

turbofan

TABLE 1-7. Typical Propulsion Characteristics of Some Rocket Applications

3 00

, O

6 o
■ 00 ^
o 2
a.

a. -

ao lo

c/5 D

•c

2 c/5

O O 0) c/5

-D

^ -5

^ .a

flj ^

^ .s

D. c/!l
c/5 flJ

Cu 6D

'5b

•c

C ■“

'c^

'3 00

.ti 1)

5 .2

^ C/5

'5 ^

O' o

e — ^

3 3 —

O O
C/5 C/5

3 '3

3 O' "O

2 3

00 c/5

O o

^ 2
C gj
3

1j o

0“ S

, o i2

Q. U c/5

o a -3
c o 5

o >- S'

S s

3 2
o s

.3 c/5

. 3 O

p (S

3 Q. C

3 O'

^ ;3

'O = =

o- 2 c

0) o
00 g
fa Xi

§■■2
d G
c/5 <

3 .3 3 Q

3 ^

'go is acceleration of gravity at the Earth’s surface = 9.8066 m/sec^ or 32.17 ft/sec^.

26 CLASSIFICATION

The term surface launch can mean a launch from the ground, the ocean surface
(from a ship), or from underneath the sea (submarine launch). Some tactical
missiles, such as the air-to-surface short-range attack missile (SRAM), have a
two-pulse solid propellant motor, where two separate, insulated grains are in the
same motor case; the time interval before starting the second pulse can be timed to
control the flight path or speed profile. Most countries now have tactical missiles
in their military inventories, and many of these countries have a capability to
produce their own rocket propulsion systems that are used to propel them.

Solid propellant rocket motors are being used today in most tactical missiles,
such as in surface-to-air, air-to-air, air-to-surface, and surface-to-surface appli-
cations, for the ejection of pilot aircraft seats or crew capsules, target drones,
signal rockets, weather sounding rockets, antitank rockets, or for the separations
of stages in a multistage flight vehicle.

Applications, which were popular 30 to 60 years ago, but are no longer active,
include liquid propellant rocket engines for propelling military fighter aircraft,
assisted takeoff rocket engines and rocket motors, and superperformance rocket
engines for augmenting the thrust of an aircraft jet engine.

REFERENCES

1-1. E. C. Goddard and G. E. Pendray (Eds), The Papers of Robert H. Goddard,
three volumes, McGraw-Hill Book Company, 1970, 1707 pages. It includes
the pioneering treatise “A Method of Reaching Extreme Altitudes” originally
published as Smithsonian Miscellaneous Collections, Vol. 71, No. 2, 1919.

1-2. G. P. Sutton, History of Liquid Propellant Rocket Ensines, published by AIAA,
2006, 911 pages.

1-3. B. N. Yur’yev (Ed), Collected Works of K. E. Tsiolkowski, Vols. 1-3, USSR
Academy of Sciences, 1951; also NASA Technical Translation F-236, April 1965.

1 -4. H. Oberth, Die Rakete zu den Planet enrdumen (By Rocket to Planetary Space),
R. Oldenburg, Munich, 1923 (in German), a classical text.

1-5. E. Sanger, Raketenflugtechnik (Rocket Flight Technology), R. Oldenburg, Munich,
1923 (in German).

1-6. W. von Braun and F. Ordway, History of Rocketry and Space Travel, 3rd ed.,
Thomas Y. Crowell, New York, 1974.

1-7. L. H. Caveny, R. L. Geisler, R. A. Ellis, and T. L. Moore, “Solid Enabling
Technologies and Milestones in the USA,” Journal of Propulsion and Power, Vol.
19, No. 6, Nov.-Dee. 2003, AIAA, pp. 1038-1066.

1-8. A. M. Lipanov, “Historical Survey of Solid Propellant Rocket Development in
Russia,” Journal of Propulsion and Power, Vol. 19, No. 6. Nov. -Dec. 2003, pp.
1063-1088.

1-9. AGARD Lecture Series 150, Design Methods in Solid Rocket Motors,
AGARD/NATO, Paris, April 1988.

1-10. A. Davenas, Solid Rocket Propulsion Technology , Pergamon Press, London (orig-
inally published in French), 1988.

REFERENCES 27

1-11. S. V. Gunn and C. M. Ehresman, “The Space Propulsion Technology Base
Established Four Decades Ago for the Thermal Nuclear Rocket Is Ready for
Current Application,” AIAA paper 2003-4590, July 2003.

1-12. R. W. Bussard and R. D. DeLauer, Nuclear Rocket Propulsion, McGraw-Hill
Book Company, New York, 1958.

1-13. D. Buden, “Nuclear Rocket Safety,” Acta Astronautica, Vol. 18, 30 Years of
Progress in Space, 1988, pp. 217-224.

1-14. R. C. Finke (Ed), Electric Propulsion and Its Application to Space Missions,
Vol. 79, Progress in Aeronautics and Astronautics Series, AIAA 1981.

1-15. R. G. Jahn, Physics of Electrical Propulsion, McGraw-Hill Book Company, New
York, 1968, 339 pages.

1-16. T. Svitek et al., “Solar Sails as Orbit Transfer Vehicle — Solar Sail Concept Study,”
Phase II Report, AIAA paper 83-1347, 1983.

1-17. V. P. Ageev et al., “Some Characteristics of the Laser Multi-pulse Explosive Type
Jet Acta Astronautica, Vol. 8, No. 5-6, 1981, pp. 625-641.

1-18. R. A. Hyde, “A Laser Fusion Rocket for Interplanetary Propulsion,” Preprint
UCRL 88857, Lawrence Livermore National Laboratory, Livermore, California,
Sept. 1983.

1-19. NASA International Space Station (A resource on the ISS by NASA; includes
operational use, wide range of background material, archives, image gallery and
planned missions) www.nase.gov/station-34k.

1-20. T. D. Thompson (Ed), TRW Space Log , Vols. 3 1 to 34, TRW Space and Electronics
Group (today part of Northrop Grumman Corp.), Redondo Beach, CA, 1996 and
1997-1998.

1-21. S. J. Isakowitz, J. B. Hopkins, and J. P. Hopkins, International Reference Guide
to Space Launch Systems, 4th ed., AIAA, 2004, 596 pages.

1 -22. National Aeronautics and Space Administration (NASA), National Space Trans-
portation System Reference, Vol. 1, Systems and Eacilities, U.S. Government
Printing Office, Washington, DC, June 1988; a description of the Space Shuttle.

1-23. A. R. Curtis (Ed), Space Satellite Handbook, 3rd ed, Gulf Publishing Company,
Houston, TX, 1994, 346 pages.

1-24. G. P. Sutton, History of Small Liquid Propellant Thrusters, presented at the 52nd
JANNAF Propulsion Meeting, May 2004, Las Vegas, NE, published by the Chem-
ical Propulsion Information Analysis Center, Columbia, Maryland, June 2004.

1-25. M. M. Micci and A. D. Ketsdever, Micropropulsion for Small Spacecraft, Progress
in Aeronautics and Astronautics Series, Vol. 187, AIAA, 2000, 477 pages.

CHAPTER 2

DEFINITIONS AND FUNDAMENTALS

The basic principles of rocket propulsion are essentially those of mechanics,
thermodynamics, and chemistry. Propulsion is achieved by applying a force to
a vehicle, that is, accelerating the vehicle or, alternatively, maintaining a given
velocity against a resisting force. This propulsive force is obtained by ejecting
propellant at high velocity. This chapter deals with the definitions and the basic
relations of this propulsive force, the exhaust velocity, and the efficiencies of
creating and converting the energy and other basic parameters. The symbols
used in the equations are defined at the end of the chapter. Wherever possible
the American Standard letter symbols for rocket propulsion (as given in Ref.
2-1) are used.

2.1. DEFINITIONS

The total impulse It is the thrust force F (which can vary with time) integrated
over the burning time t:

For constant thrust and very short start and stop transients this reduces to

Impulse If is proportional to the total energy released by all the propellant in a
propulsion system.

( 2 - 1 )

It = Ft

( 2 - 2 )

2.1. DEFINITIONS 29

The specific impulse is the total impulse per unit weight of propellant. It is
an important figure of merit of the performance of a rocket propulsion system,
similar in concept to the miles per gallon parameter used with automobiles.
A higher number means better performance. Values of Is are given in many
chapters of this book, and the concept of an optimum specific impulse for a
particular mission is introduced later. If the total mass flow rate of propellant is
m and the standard acceleration of gravity at sea level go is 9.8066 m/sec^ or
32.174 ft/sec^, then

IpFdt

gojrhdt

(2-3)

This equation will give a time-averaged specific impulse value for any rocket
propulsion system, particularly where the thrust varies with time. During tran-
sient conditions (during start or the thrust buildup period, the shutdown period,
or during a change of flow or thrust levels) values of /* can be obtained by inte-
gration or by determining average values for F and m for short time intervals.
For constant thrust and propellant flow this equation can be simplified; below,
Mp is the total effective propellant mass expelled through the nozzle.

h = It/impgo) (2-4)

In Chapter 3 there is further discussion of the specific impulse. For constant
propellant mass flow m, constant thrust F, and negligibly short start or stop
transients:

Is = F/(mgo) = F/w

hKmpgo) = It/w (2-5)

at or near the earth’s surface, the product ffipgo is the total effective propellant
weight w, and the weight flow rate is w. But in space or outer satellite orbits,
the mass is just multiplied by an arbitrary constant, namely go- In the Systeme
International (SI) or metric system of units 4 can be expressed simply in “sec-
onds” because of the use of the constant go- In the United States today we still
use the English Engineering (EE) system of units (foot, pound, second) in many
of the chemical propulsion engineering, manufacturing, and test operations. In
many past and current U.S. publications, data, and contracts, the specific impulse
has units of thrust (Ibf) divided by weight flow rate of propellants (Ibf/sec),
simplified as seconds. The numerical value of 4 is the same in the EE and
the SI system of units. However, the units of Is do not represent a measure of
elapsed time but a thrust force per unit “weight” flow rate. In this book the
symbol Is is used for the specific impulse, as listed in Ref. 2-1. Eor solid pro-
pellant and several other rocked systems the symbol fip is often used, as listed in
Ref. 2-2.

30 DEFINITIONS AND FUNDAMENTALS

In a rocket nozzle the actual exhaust velocity is not uniform over the entire
exit cross section and does not represent the entire thrust magnitude. The velocity
profile is difficult to measure accurately. For convenience a uniform axial velocity
c is assumed, which allows a one-dimensional description of the problem. This
ejfective exhaust velocity c is the average equivalent velocity at which propellant
is ejected from the vehicle. It is defined as

C = hgQ = F /ih (2-6)

It is given either in meters per second or feet per second. Since c and /j differ
only by an arbitrary constant, either one can be used as a measure of rocket
performance. In the Russian literature c is generally used in lieu of /* .

In solid propellant rockets it is difficult to measure the propellant flow rate
accurately. Therefore, the specific impulse is often calculated from total impulse
and the propellant weight (using the difference between initial and final motor
weights and Eq. 2-5). In turn the total impulse is obtained from the integral of the
measured thrust with time, using Eq. 2-1. In liquid propellant units it is possible
to measure thrust and instantaneous propellant flow rate and thus to use Eq. 2-3
for calculation of specific impulse. Equation 2-4 allows another definition for
specific impulse, namely, fhe amount of impulse imparted to a vehicle per unit
sea-level weight of propellant expended.

The term specific propellant consumption refers to the reciprocal of the specific
impulse and is not commonly used in rocket propulsion. It is used in automotive
and duct propulsion systems. Typical values are listed in Table 1-2.

The mass ratio IVR of a vehicle or a particular vehicle stage is defined to
be the final mass mj (after rocket operation has consumed all usable propel-
lant) divided by mo (before rocket operation). The various terms are depicted in
Fig. 4-1.

IVR = m//mo (2-7)

This applies to a single or a multistage vehicle; for the latter, the overall mass
ratio is the product of the individual vehicle stage mass ratios. The final mass
m/ is the mass of the vehicle after the rocket has ceased to operate when all
the useful propellant mass nip has been consumed and ejected. The final vehicle
mass mf includes all those components that are not useful propellant and may
include guidance devices, navigation gear, payload (e.g., scientific instruments
or a military warhead), flight control systems, communication devices, power
supplies, tank structure, residual or unusable propellant, and all the propulsion
hardware. In some vehicles it can also include wings, fins, a crew, life support
systems, reentry shields, landing gears, and the like. Typical values of IVR can
range from 60% for some tactical missiles to less than 10% for some unmanned
launch vehicle stages. This mass ratio is an important parameter in analyzing
flight performance, as explained in Chapter 4. When IVR is applied to a single
stage, then its upper stages become the “payload.”

2.1. DEFINITIONS 31

The propellant mass fraction ^ indicates the fraction of propellant mass
in an initial mass mq. It can he applied to a vehicle, a stage of a vehicle, or to a
rocket propulsion system.

^ = Mp/mo ( 2 - 8 )

C = (mo - mf)lmQ = mp/{mp + mj) (2-9)

mo = mf + lUp ( 2 - 10 )

When applied to a rocket propulsion system, the mass ratio IVR and propellant
fraction (; are different from those that apply to a vehicle as described above.
Here the initial or loaded mass mo consists of the inert propulsion mass (the
hardware necessary to burn and store the propellant) and the effective propellant
mass. It would exclude masses of nonpropulsive components, such as payload or
guidance devices. For example, in a liquid propellant rocket engine the final or
inert propulsion mass m/ would include the propellant feed tanks, their pressur-
ization system (with turbopump and/or gas pressure system), one or more thrust
chambers, various piping, fittings and valves, an engine mount or engine struc-
ture, filters, and some sensors. The residual or unusable remaining propellant
is usually considered to be part of the final inert mass mf, as it will be in this
book. However, some rocket propulsion manufacturers and some literature assign
residuals to be part of the propellant mass mp . When applied to a rocket propul-
sion system, the value of the propellant mass fraction indicates the quality of
the design; a value of, say, 0.91 means that only 9% of the mass is inert rocket
hardware, and this small fraction contains, feeds, and burns a substantially larger
mass of propellant. A high value of f is desirable.

The impulse-to-weight ratio of a complete propulsion system is defined as the
total impulse f divided by the initial or propellant-loaded vehicle weight wq. A
high value indicates an efficient design. Under our assumptions of constant thrust
and negligible start and stop transients, it can be expressed as

Wo (m/ -F mp)go
=

mf/mp + 1

( 2 - 11 )

( 2 - 12 )

The thrust-to-weight ratio F /wq expresses the acceleration (in multiples of
the Earth’s surface acceleration of gravity) that the engine is capable of giving
to its own loaded propulsion system mass. For constant thrust the maximum
value of the thrust-to-weight ratio, or maximum acceleration, occurs just before
termination or burnout because the vehicle mass has been diminished by the mass
of useful propellant. Values of F /w are given in Table 2-1. The thrust-to-weight
ratio is useful to compare different types of rocket systems.

TABLE 2-1. Ranges of Typical Performance Parameters for Various Rocket Propulsion Systems

3 1

.3 BO

D. C

“ I

Q. O
c^CIh

a _

.2 §

Oh U

t-H

OhO

^ 3 D

^ CIh

*ob

c c

(D (U
> >
o o

s D « -H
g- Oh Oh Oh

5 G-

o o x: 4=

> 5P op

g :

o c

:9 3 5

5 "o

!3r tA Gh ^

pH PP

^ E

Z |_3’ <N

. X E

<N (N

X z

D D <N

E X Z E

r «~)

— H

"-H

— H

"-H

Weeks

few

'§

C/5

Days

4 Ch

Days

Tl -

o '

r -

I >

o \

1-H

I >

4—1

x :

’o

1 j

^ ,

’ Gh

f -,

’2

G -

-g

4 =

-3 ,D
D

E ^

4 =

O "

3 _i

- E3

Ion -

’o

G 3

CS D
C ^

.2 e

CA D

Q. S'

D C

3 ^

s ^

o ' ^

7D O

'cA t2

^ II

o ^
^ 2
o ^

0.-S

< Qi

Kinetic power per unit exhaust mass flow.

2.2. THRUST 33

Example 2-1. A rocket projectile has the following characteristics:

Initial mass

Mass after rocket operation
Payload, nonpropulsive structure, etc.
Rocket operation duration
Average specific impulse of propellant

200 kg
130 kg
110 kg
3.0 sec
240 sec

Determine the mass ratios for the overall vehicle and for the propulsive unit, the
propellant mass fraction for the propulsive unit, the effective exhaust velocity, and the total
impulse. Also, sensitive electronic equipment in the payload would limit the maximum
acceleration to 35go’s. Is this exceeded during the flight? Assume constant thrust and
neglect the short start and stop transients.

SOLUTION. From the givens, the mass of the expended propellant is 70 kg. The mass
ratio (Eq. 2-8) for the vehicle is therefore IVR = mf/mo = 130/200 = 0.65, and the the
propulsion unit is (130 — 110)/(200 — 110) = 0.222. Note that these are different. Now
the propellant mass fraction of the propulsion system is

Note that this mass fraction is acceptable for tactical missiles but might be only “fair”
for spacecraft. Because the effective exhaust velocity is proportional to the specific
impulse, c = 240 x 9.81 = 2354.4 m/sec. The propellant mass flow rate is m = 70/3 =
23.3 kg/sec and the thrust becomes (Eq. 2-5)

Now the total impulse can be calculated as /, = 54936 x 3 = 164,808 N-sec. The
impulse-to- weight ratio (Eq. 2-11) for the propulsion system at the earth’s surface is
I,/wq = 164,808/[(200 — 110)9.81] = 187 sec (this reflects only a fair design because
Is = 240 sec).

For a horizontal trajectory, the maximum acceleration is found at the end of the
thrusting schedule just before shutdown (because while the thrust is unchanged the mass
is now at its minimum value of 130 kg).

This represents 43.08 go’s, at the earth’s surface and exceeds the limit of the
equipment.

2.2. THRUST

The thrust is the force produced by a rocket propulsion system acting upon a
vehicle. In a simplified way, it is the reaction experienced by its structure due to
the ejection of matter at high velocity. It represents the same phenomenon that

C = 70/(200- 110) = 0.778

F = 23.3 X 2354.4 = 54,936 N

Final acceleration = F fm = 54,936/130 = 422.58 m/sec^

34 DEFINITIONS AND FUNDAMENTALS

pushes a garden hose backwards or makes a gun recoil. In the latter case, the
forward momentum of the bullet and the powder charge is equal to the recoil
or rearward momentum of the gun barrel. Momentum is a vector quantity and
is defined as the product of mass times velocity. All ship propellers and oars
generate their forward push at the expense of the momentum of the water or air
masses, which are accelerated toward the rear. Rocket propulsion differs from
these devices primarily in the relative magnitude of the accelerated masses and
velocities. In rocket propulsion relatively small masses are involved that are
carried within the vehicle and ejected at high velocities.

The thrust, due to a change in momentum, is given below. A derivation can be
found in earlier editions of this book. The thrust and the mass flow are constant
and the gas exit velocity is uniform and axial.

dm w

F = —V2 = mv2 = —V2 (2-13)

dt go

This force represents the total propulsion force when the nozzle exit pressure
equals the ambient pressure.

The pressure of the surrounding fluid (i.e., the local atmosphere) gives rise to
the second contribution that influences the thrust. Figure 2-1 shows schematically
the external pressure acting uniformly on the outer surface of a rocket chamber
and the changing gas pressures on the inside of a typical thermal rocket engine.
The size and length of the arrows indicates the relative magnitude of the pressure

Atmosphere Converging nozzle section

FIGURE 2-1. Pressure balance on chamber and nozzle interior walls is not uniform. The
internal gas pressure (indicated by length of arrows) is highest in the chamber (pi) and
decreases steadily in the nozzle until it reaches the nozzle exit pressure p 2 - The external
or atmospheric pressure p 3 is uniform. At the throat the pressure is /?,. The four subscripts
(shown inside circles) refer to the quantities A, v, T, and p at specific locations.

2.2. THRUST 35

forces. The axial thrust can be determined by integrating all the pressures acting
on areas that can be projected on a plane normal to the nozzle axis. The forces
acting radially outward are appreciable but do not contribute to the axial thrust
because a rocket is typically an axially symmetric chamber. The conditions prior
to entering the nozzle are essentially stagnation conditions.

Because of a fixed nozzle geometry and changes in ambient pressure due to
variations in altitude, there can be an imbalance of the external environment
or atmospheric pressure and the local pressure p 2 of the hot gas jet at the
exit plane of the nozzle. Thus, for a steadily operating rocket propulsion system
moving through a homogeneous atmosphere, the total thrust is equal to (this
equation is arrived at using the control volume approach in gas dynamics; see
Refs. 2-3 and 2-4);

F = mv2 + (P2 - Pi)A2 (2-14)

The first term is the momentum thrust represented by the product of the pro-
pellant mass flow rate and its exhaust velocity relative to the vehicle. The second
term represents the pressure thrust consisting of the product of the cross-sectional
area at the nozzle exit A 2 (where the exhaust jet leaves the vehicle) and the differ-
ence between the exhaust gas pressure at the exit and the ambient fluid pressure.
If the exhaust pressure is less than the surrounding fluid pressure, the pressure
thrust is negative. Because this condition gives a low thrust and is undesirable,
the rocket nozzle is usually so designed that the exhaust pressure is equal or
slightly higher than the ambient fluid pressure.

When the ambient atmosphere pressure is equal to the exhaust pressure, the
pressure term is zero and the thrust is the same as in Eq. 2-13. In the vacuum
of space P 3 =0 and the thrust becomes

F = mv2+P2Ai (2-15)

At the altitude where the exhaust pressure is exactly equal to the surrounding
fluid pressure (p 2 = P$) the pressure thrust is zero. For any nozzle this occurs
only at one attitude, and this is referred to as the rocket nozzle with optimum
nozzle expansion ratio A 2 /A;. This is further elaborated in Chapter 3.

Equation 2-14 shows that the thrust of a rocket unit is independent of the
flight velocity. Because changes in ambient pressure affect the pressure thrust,
there is a variation of the rocket thrust with altitude. Because atmospheric pressure
decreases with increasing altitude, the thrust and the specific impulse will increase
as the vehicle is propelled to higher altitudes. This change in pressure thrust due
to altitude changes can amount to between 10 and 30% of the overall thrust, as is
shown for a typical rocket engine in Fig. 2-2. Table 8-1 shows the sea level and
high-altitude thrust for several rocket engines. Appendix 2 gives the properties
of the standard atmosphere (ambient pressure p^).

36 DEFINITIONS AND FUNDAMENTALS

FIGURE 2-2. Altitude performance of RS 27 liquid propellant rocket engine used in
early versions of the Delta launch vehicle.

2.3. EXHAUST VELOCITY

The ejfective exhaust velocity as defined by Eq. 2-6 applies to all rockets that
thermodynamically expand hot gas in a nozzle and, indeed, to all mass expul-
sion systems. From Eq. 2-14 and for constant propellant mass flow this can be
modified to

c = V2 + {p2- Pz)^ilm ( 2 - 16 )

Equation 2-6 shows that c can be determined from thrust and propellant flow
measurements. When p 2 = Pi, the effective exhaust velocity c is equal to the
average actual exhaust velocity of the propellant gases V 2 - When p 2 ^ Pi, then
c zfz V 2 - The second term of the right-hand side of Eq. 2-16 is usually small
in relation to V 2 , thus the effective exhaust velocity is usually close in value to
the actual exhaust velocity. When c = V 2 , the thrust (from Eq. 2-14) can be
rewritten as

F = (w I go)v 2 = the (2-17)

The characteristic velocity has been used frequently in the rocket propulsion
literature. Its symbol c*, pronounced “cee-star,” is defined as

c*=piAt!m (2-18)

The characteristic velocity c* is used in comparing the relative performance of
different chemical rocket propulsion system designs and propellants; it is easily
determined from measured data of m, p\, and At. It relates to the efficiency of
the combustion and is essentially independent of nozzle characteristics.

2.3. EXHAUST VELOCITY 37

However, the specific impulse 4 and the effective exhaust velocity c are
functions of the nozzle geometry, such as the nozzle area ratio A 2 /A,, as shown
in Chapter 3. Some values of 4 and c* are given in Tables 5-5 and 5-6.

Example 2-2. The following measurements were made a sea-level test of a solid pro-
pellant rocket motor (all cross sections are circular and unchanging):

Bum duration

40 sec

Initial propulsion system mass

1210 kg

Mass of rocket motor after test

215 kg

Sea-level thrust

62,250 N

Chamber pressure

7.00 MPa

Nozzle exit pressure

70.0 kPa

Nozzle throat diameter

8.55 cm

Nozzle exit diameter

27.03 cm

Determine m, V 2 , c* , and c at sea level. Also, determine the pressure thrust and specific
impulse at sea level, 1000 m, and 25,000 m altitude. Assume the momentum thrust is
invariant during the ascent, and the start and stop transients are short.

SOLUTION. The cross sections corresponding to the given diameters are A, = 0.00574
and A 2 = 0.0574 m^ (i.e., an area ratio of 10). The mass flow rate for all altitudes is

m = (1210 - 215)/40 = 24.88 kg/sec

The desired c* (all altitudes) and c at sea level follow (Eqs. 2-6 and 2-18):

c* = piA,lm = 7.00 X 106 x 0.00574/24.88 = 1615 m/sec
c = F/in = 62250/24.88 = 2502 m/sec

The pressure thrust at sea level is

(pi -P3)A2 = (0.070 - 0.1013) X 10® X 0.0574 = -1797 N
so the nozzle exit velocity becomes (all altitudes), from Eq. 2-14,

V 2 = (62250 - 1797)/24.88 = 2430 m/sec

The remaining answers are shown in the table below. We obtain the ambient pressures
from Appendix 2.

Altitude

P 3 (kPa)

Pressure Thrust (N)

4 (sec)

Sea level

101.32

-1798

248

1000 m

89.88

-1141

250

25,000 m

2.55

3871

111

Further examination of the standard atmosphere table in Appendix 2 reveals that the
pressure thrust becomes zero at an elevation of just under 3000 m.

38 DEFINITIONS AND FUNDAMENTALS

2.4. ENERGY AND EFFICIENCIES

Although efficiencies are not commonly used directly in designing rocket units,
they permit an understanding of the energy balance of a rocket system. Their
definitions are arbitrary, depending on the losses considered, and any consistent
set of efficiencies, such as the one presented in this section, is satisfactory in
evaluating energy losses. As stated previously, two types of energy conversion
processes occur in any propulsion system, namely, the generation of energy,
which is really the conversion of stored energy into available energy and, subse-
quently, the conversion to the form in which a reaction thrust can be obtained.
The kinetic energy of ejected matter is the form of energy useful for propulsion.
The power of the jet Pjet is the time rate of expenditure of this energy, and for a
constant gas ejection velocity v this is a function of f and P:

The term specific power is sometimes used as a measure of the utilization
of the mass of the propulsion system including its power source; it is the jet
power divided by the loaded propulsion system mass, Pjet /mo. For electrical
propulsion systems that carry a heavy, relatively inefficient energy source, the
specific power can be much lower than that of chemical rockets. The energy input
from the energy source to the rocket propulsion system has different forms in
different rocket types. For chemical rockets the energy is created by combustion.
The maximum energy available per unit mass of chemical propellants is the heat
of the combustion reaction Qr ; the power input to a chemical engine is

where 7 is a conversion constant that depends on the units used. A large portion of
the energy of the exhaust gases is unavailable for conversion into kinetic energy
and leaves the nozzle as residual enthalpy. This is analogous to the energy lost
in the hot exhaust gases of internal combustion engines.

The combustion efficiency for chemical rockets is the ratio of the actual and
the ideal heat of reaction per unit of propellant and is a measure of the source
efficiency for creating energy. Its value is high (approximately 94 to 99%), and it
is defined in Chapter 3. When the power input Pchem is multiplied by the combus-
tion efficiency, it becomes the power available to the propulsive device, where it
is converted into the kinetic power of the exhaust jet. In electric propulsion the
analogous efficiency is the power conversion efficiency. For solar cells it has a
low value; it is the efficiency for converting solar radiation energy into electric
power (10 to 20%).

The power transmitted to the vehicle at any one time is defined in terms of
the thrust of the propulsion system F and the vehicle velocity u :

Fjet = jthv^ = jWgol^ — jFgof = jFV2

(2-19)

^chem —

( 2 - 20 )

^vehicle —

(2-21)

2.4. ENERGY AND EFFICIENCIES 39

The internal efficiency of a rocket propulsion system is an indication of the effec-
tiveness of converting the system’s energy input to the propulsion device into
the kinetic energy of the ejected matter; for example, for a chemical unit it is the
ratio of the kinetic power of the ejected gases expressed by Eq. 2-19 divided by
the power input of the chemical reaction as given in Eq. 2-20. Internal efficien-
cies are used in Example 2-4. The energy balance diagram for a chemical rocket
(Eig. 2-3) shows typical losses. The internal efficiency can be expressed as

'lint =

kinetic power in jet
available chemical power

'Icomb^chem

( 2 - 22 )

Typical values of jjint are listed later in Example 2-4.

An object moving through a fluid affects the fluid (i.e., stirs it) in ways that
may hinder its motion and/or require extra energy expenditures. This is one
consequence of skin friction and can be substantial. The propulsive efficiency
(Eig. 2-4) reflects this energy cost for rocket vehicles. The equation that deter-
mines how much of the kinetic energy of the exhaust jet is useful for propelling
a rocket vehicle is defined as

vehicle power

rjp =

vehicle power -f residual kinetic jet power
Fu 2u fc

= = (2-23)

Em - h im(c - m)^ 1 -f- ( m/c)^

where F is the thrust, u the absolute vehicle velocity, c the effective rocket
exhaust velocity with respect to the vehicle, m the propellant mass flow rate, and
rjp the propulsive efficiency. The propulsive efficiency is a maximum when the

Heat loss to walls

Combustion loss
(poor mixing,
incomplete burning)

100%

Unavailable thermal
energy of exhaust jet

Residual kinetic energy
of exhaust gases
^0to50%

40 to 70%

LUseful energy for
vehicle propulsion
— Kinetic energy of exhaust jet
-Total energy of exhaust jet
-Available energy in combustion chamber
-Heating value of propellants

FIGURE 2-3. Typical energy balance diagram for a chemical rocket.

40 DEFINITIONS AND FUNDAMENTALS

0 1.0 2.0 3.0

Velocity ratio, u/c

FIGURE 2-4. Propulsive efficiency at varying velocities.

forward vehicle velocity is exactly equal to the exhaust velocity. Then the residual
kinetic energy and the absolute velocity of the jet are zero and the exhaust gases
stand still in space.

While it is desirable to use energy economically and thus have high efficien-
cies, there is also the problem of minimizing the expenditure of ejected mass,
which in many cases is more important than minimizing the energy. In nuclear
reactor energy and some solar energy sources, for example, there is an almost
unlimited amount of heat energy available; yet the vehicle can only carry a limited
amount of working fluid. Economy of mass expenditures of working fluid can be
obtained if the exhaust velocity is high. Because the specific impulse is propor-
tional to the exhaust velocity, it is a measure of this propellant mass economy.

2.5. MULTIPLE PROPULSION SYSTEMS

The relationships below are for determining the overall (oa) or total thrust and the
overall mass flow of propellants for a group of propulsion systems (two or more)
bring in parallel (i.e., in the same direction at the same time). These relationships
apply to liquid propellant rocket engines, solid propellant rocket motors, electrical
propulsion systems, hybrid propulsion systems, and to any combinations of these.
Many space launch vehicles and larger missiles have multiple propulsion systems.
The Space Shuttle, for example, has three larger liquid engines and two large
solid motors bring jointly at liftoff.

The overall thrust, Eoa, is needed for determining the bight path, and the overall
mass bow rate, moa, is needed for determining the vehicle’s mass decrease with
time. Together these two determine the overall specibc impulse, iIs)on'

= = Fi+F2 + F3 + ... (2-24)

h?oa = = nil + m2 + m + . . . (2-25)

(4)oa = ^ F/(go ^ m)

( 2 - 26 )

2.6. TYPICAL PERFORMANCE VALUES 41

For liquid propellant rocket engines with a turbopump and a gas generator,
there is a separate turbine outlet flow that is usually dumped overboard through
a pipe and a nozzle (see Fig. 1 -4), which needs to be included in the equations
above.

Example 2-3. The MA-3 multiple liquid engine for the Atlas missile has two booster
engines (F = 165,000 Ibf at sea level and m = 66.71bm/sec each). The turbine exhaust
gas of both boosters provides an extra 2300 Ibf of thrust at a propellant flow rate of
about 3.3 Ibm/sec. These booster engines are dropped from the vehicle after about 145
sec operation. The central sustainer rocket engine, which also starts at liftoff, continues
to operate for 300 sec. It has a sea-level thrust of 57{,j000 Ibf at a mass flow of 27.05
Ibm/sec. Its turbine exhaust gases are aspirated into the nozzle of the engine. There are
also two small vernier engines used for roll control of the vehicle during the sustainer-only
portion of the flight. They each have a thrust of 415 Ibf at sea level with a propellant flow
of 0.213 Ibm/sec. Determine the overall thrust Foa and the overall mass flow rate moa at
liftoff, when all the nozzles are pointing vertically down.

SOLUTION. Use Eqs. 2-24 and 2-25. The thrust from the turbine must be included in
2-24.

Foa = 165,000 X 2 -h 2300 x 2 -h 57,000 -h 415 x 2 = 390,430 Ibf

moa = 66.7 X 2 -h 3.3 X 2 -h 27.05 -h 0.213 x 2 = 167.5 Ibm/sec

2.6. TYPICAL PERFORMANCE VALUES

Typical values of representative performance parameters for different types of
rocket propulsion are given in Table 2-1 and in Fig. 2-5.

Chemical rockets have relatively low values of specific impulse, relatively light
machinery (i.e., low engine weight), a very high thrust capability, and therefore
high acceleration and high specific power. At the other extreme, the ion propul-
sion devices have a very high specific impulse, but they must carry a heavy
electrical power source with them to deliver the power necessary for high ejec-
tion velocities. The very low acceleration potential for the electrical propulsion
units and those using solar radiation energy usually requires a long period for
accelerating and thus these systems are best used for missions where the flight
time is long. The low thrust values of electrical systems imply that they are not
useful in fields of strong gravitational gradients (for takeoff or landing) but are
best used in a true space flight mission.

The chemical systems (solid and liquid propellant rockets) are fully developed
and widely used for many different vehicle applications. They are described in
Chapters 5 to 16. Electrical propulsion has been in operation in many space
flight applications (see Chapter 17). Some of the other types are still in their
exploratory or development phase, but may become useful.

Example 2-4. Compare the types of propulsion system tabulated below by computing
the power input and propellant flow required for each to deliver 106.8 N (25 Ibf) of thrust.

42 DEFINITIONS AND FUNDAMENTALS

10 ® 10 ® 10 '' 10 ® 10 2 10 1 1 10 100 1000
Acceleration in multiples of earth gravity Qq or thrust to
vehicle \«eight ratio

FIGURE 2-5. Exhaust velocities as a function of typical vehicle accelerations. Regions
indicate approximate performance values for different types of propulsion systems. The
mass of the vehicle includes the propulsion system, but the payload is assumed to be
zero.

SOLUTION. From Eqs. 2-13 and 2-19 and assuming that V 2 = c.

m = F/I,go

Power input = Pjet/hint = V2 mvllmm

Engine Type

hint

Is (sec)

V 2 (m/sec)

in (kg/sec)

Pinput(kW)

Chemical rocket

0.70

300

2,943

[0.0363]

[225]

Nuclear fission

0.50

800

7,848

[0.0136]

[838]

Arc-electrothermal

0.40

600

5,886

[0.0181]

[785]

Ion electrostatic

0.65

2000

19,620

[0.0054]

[1612]

Results are shown inside square brackets in the table. More than half a megawatt input is
needed for the last three propulsion systems but note that the propellant flows are small.
These data are only illustrative because to date there are no single arc or ion thrusters
that supply the required thrust so multiple units would be required. Electric propulsion is
presently restricted to thrust levels of less than one newton in part because of the lack

PROBLEMS 43

of such large electrical power supplies. Nuclear propulsion units are not available at the
present time so chemical propulsion would remain as the only option. Note that it would
be more realistic to compare total impulse because thrust times can vary between types.

Variable Thrust

There are flight missions that require a change in thrust during flight. Some solid
rocket propellant motors can have a fixed predetermined thrust usually consisting
of an initial short duration high-thrust boost phase followed by a lower thrust
longer sustaining phase. Details may be found in Section 12.3. Some liquid
propellant rocket engines can have random variable thrust that can be adjusted
during flight. Details are in Section 8.7. In small low-thrust liquid propellant
attitude control thrusters, it is possible to control the number and/or timing of
repeated short pulses to achieve a variable thrust effect. This concept is discussed
in Section 8.3.

A predetermined thrust decrease (used in some solid propellant rocket motors)
can be obtained with a special design of the grain allowing a high-thrust period
(for missile acceleration) followed by a low-thrust period for maintaining the
flight velocity against atmospheric drag. Alternatively, it may be a special two-
pulse motor with two different grain compositions in the same combustion cam-
ber, one for a high-thrust boost phase and one for a subsequent low-thrust
sustaining phase. Both of these are discussed in Chapter 12.

Some solid propellant and liquid propellant experimental propulsion systems
have used variable nozzle throat areas (achieved with a variable position “tapered
pintle” at the nozzle throat) and one experimental version has flown. To date, there
has been no published information on the further production and implementation
of such systems.

PROBLEMS

When solving problems, three appendixes (see end of book) may be helpful:

Appendix 1. Conversion Factors and Constants
Appendix 2. Properties of the Earth’s Standard Atmosphere
Appendix 3. Summary of Key Equations

1 . A jet of fluid hits a stationary flat plate in the manner shown below.

44 DEFINITIONS AND FUNDAMENTALS

(a) If there is 50 kg of fluid flowing per minute at an absolute velocity of 200 m/sec,
what will be the force on the plate?

Answer: 167 N.

(b) What will this force be when the plate moves in the direction of flow at n = 50
km/h? Explain the methodology.

Answer: 144 N.

2. The following data are given for a certain rocket unit: thrust, 8896 N; propellant
consumption, 3.867 kg/sec; velocity of vehicle, 400 m/sec; energy content of propellant,
6.911 Ml/kg. Assume 100% combustion efficiency.

Determine (a) the effective velocity; (b) the kinetic jet energy rate per unit flow of
propellant; (c) the internal efficiency; (d) the propulsive efficiency; (e) the overall
efficiency; (f) the specific impulse; (g) the specific propellant consumption.

Answers: (a) 2300 m/sec; (b) 2.645 MJ/kg; (c) 38.3%; (d) 33.7%; (e) 13.3%;
(f) 234.7 sec; (g) 0.00426 sec”'.

3. A certain rocket has an effective exhaust velocity of 7000 ft/sec; it consumes 280
Ibm/sec of propellant mass, each of which liberates 2400 Btu/lbm. The unit operates
for 65 sec. Construct a set of curves plotting the propulsive, internal, and overall
efficiencies versus the velocity ratio ujciQ < ujc < 1.0). The rated flight velocity
equals 5000 ft/sec. Calculate (a) the specific impulse; (b) the total impulse; (c) the
mass of propellants required; (d) the volume that the propellants occupy if their average
specific gravity is 0.925.

Answers: (a) 217.5 sec; (b) 3,960,000 Ibf-sec; (c) 18,200 Ibm; (d) 315 ftl

4. For the rocket in Problem 2, calculate the specific power, assuming a propulsion system
dry mass of 80 kg and a duration of 3 min.

5. A Russian rocket engine (RD-110) consists of four nonmoveable thrust chambers
supplied by a single turbopump. The exhaust from the turbine of the turbopump then
drives four vernier chamber nozzles (which can be rotated to provide some control of
the flight path). Using the information below, determine the thrust, effective exhaust
velocity, and mass flow rate of the four vernier thrusters.

Individual thrust chambers (vacuum):

F = 73.14 kN,c = 3279 m/sec
Overall engine with verniers (vacuum):

F = 297.93 kN,c = 3197 m/sec

Answers: 5.358 kN, 1351 m/sec, 3.965 kg/sec.

6. A certain rocket engine has a specific impulse of 250 sec. What range of vehicle
velocities {u, in units of ft/sec) would keep the propulsive efficiencies at or greater
than 80%. Also, how could rocket-vehicle staging be used to maintain these high
propulsive efficiencies for the range of vehicle velocities encountered during launch?
Answers: 4021 to 16,085 ft/sec; design upper stages with increasing /j.

SYMBOLS 45

7. Plot the variation of the thrust and specific impulse against altitude, using the atmo-
spheric pressure information given in Appendix 2, and the data for the Minuteman
first-stage rocket thrust chamber in Table 12-3. Assume that p 2 = 8.66 psia.

8 . During the boost phase, the three Space Shuttle main engines (SSMEs) operate together
with the two solid propellant rocket motors (SRBs) for the first 2 minutes. For the
remaining thrust time, the SSMEs operate alone. Using Table 1-4, calculate the overall
specific impulse for the vehicle during the 2-minute combined thrust operation.
Answer: 310 sec.

9. For the values given in Table 2-1 for the various propulsion systems, calculate the
total impulse for a fixed propellant mass of 2000 kg.

SYMBOLS (English engineering units are given in parentheses)

A area, m^(ft^)

At nozzle throat area, m^(ft^)

A 2 exit area of nozzle, m^(ft^)

c effective velocity, m/sec (ft/sec)

c* characteristic velocity, m/sec (ft/sec)

E energy, J (ft-lbf)

F thrust force, N (Ibf)

For overall force, N (Ibf)

go standard sea-level acceleration of gravity,

9.80665 m/sec2(32.174 ft/sec^)

Is specific impulse, sec

Us) OR overall specific impulse, sec

It impulse or total impulse, N-sec (Ibf-sec)

J conversion factor or mechanical equivalent of heat, 4.184 J/cal or

1055 J/Btu or 778 ft-lbf/Btu

m mass, kg (slugs) (1 slug = mass of 32.174 lb of weight at sea

level)

nioR overall mass, kg

m mass flow rate, kg/sec (Ibm/sec)

nif final mass (after rocket propellant is ejected), kg (Ibm or slugs)

nip propellant mass, kg (Ibm or slugs)

mo initial mass (before rocket propellant is ejected), kg (Ibm or slugs)

IVR mass ratio (m/-/mo)

p pressure, pascal [Pa] or N/m^(lbf/ft^)

P 3 ambient or atmospheric pressure, Pa (Ibf/ft^)

P 2 rocket gas pressure at nozzle exit, Pa (Ibf/ft^)

Pi chamber pressure, Pa (Ibf/ft^)

P power, J/sec (ft-lbf/sec)

Pj specific power, J/sec-kg (ft-lbf/sec-lbm)

46 DEFINITIONS AND FUNDAMENTALS

Qr heat of reaction per unit propellant, J/kg (Btu/lbm)

1 time, sec

u vehicle velocity, m/sec (ft/sec)

V2 gas velocity leaving the rocket, m/sec (ft/sec)

w weight, N or kg-m/sec^ (Ibf)

w weight flow rate, N/sec (Ibf/sec)

Wo initial weight, N or kg-m/sec^ (Ibf)

Greek Letters

^ propellant mass fraction

r] efficiency

>?comb combustion efficiency

>7int internal efficiency

r]p propulsive efficiency

REFERENCES

2-1. “American National Standard Letter Symbols for Rocket Propulsion”, ASME Pub-
lication Y 10.14, 1959.

2-2. “Solid Propulsion Nomenclature Guide,” CPIA Publication 80 , Chemical Propulsion
Information Agency, Johns Hopkins University, Laurel, MD, May 1965, 18 pages.
2-3. P. G. Hill and C. R. Peterson, Mechanics and Thermodynamics of Propulsion,
Addison- Wesley, Reading, MA, 1992.

2-4. R. D. Zucker and O. Bihlarz, Fundamentals of Gas Dynamics, 2nd ed., John Wiley
& Sons, Hoboken, NJ, 2002.

CHAPTER 3

NOZZLE THEORY AND
THERMODYNAMIC RELATIONS

Thermodynamic relations of the processes inside a rocket nozzle and chamber
furnish the mathematical tools needed to calculate the performance and deter-
mine several of the key design parameters of rocket propulsion systems. They
are useful as a means of evaluating and comparing the performance of various
rocket systems; they permit the prediction of the operating performance of any
rocket unit that uses the thermodynamic expansion of a gas and the determination
of several necessary design parameters, such as nozzle size and generic shape,
for any given performance requirement. This theory applies to chemical rocket
propulsion systems (both liquid and solid and hybrid propellant types), nuclear
rockets, solar-heated and resistance or arc-heated electrical rocket systems, and
to any propulsion system that uses the expansion of a gas as the propulsive
mechanism for ejecting matter at high velocity.

These thermodynamic relations, which are fundamental and important in anal-
ysis and design of rocket units, are introduced and explained in this chapter. The
utilization of these equations should give the reader a basic understanding of
the thermodynamic processes involved in rocket gas behavior and expansion. A
knowledge of elementary thermodynamics and fluid mechanics on the part of the
reader is assumed (see Refs. 3-1, 3-2, and 3-3). This chapter also addresses
different nozzle configurations, nonoptimum performance, energy losses, noz-
zle alignment, variable thrust, and four different ways for establishing nozzle
performance parameters.

48 NOZZLE THEORY AND THERMODYNAMIC RELATIONS

3.1. IDEAL ROCKET

The concept of ideal rocket propulsion systems is useful because the relevant
basic thermodynamic principles can be expressed as simple mathematical rela-
tionships, which are given in subsequent sections of this chapter. These equations
theoretically describe a quasi-one-dimensional nozzle flow, which corresponds to
an idealization and simplification of the full two- or three-dimensional equations
and the real aerothermochemical behavior. However, with the assumptions and
simplifications stated below, they are very adequate for obtaining useful solutions
to many rocket propulsion systems and preliminary design tasks. For chemical
rocket propulsion the measured actual performance is usually between 1 and
6% below the calculated ideal value. In designing new rockets, it has become
accepted practice to use ideal rocket parameters that can then be modified by
appropriate corrections, such as those discussed in Section 3.5. An ideal rocket
unit is one for which the following assumptions are valid:

1. The working substance (or chemical reaction products) is homogeneous .

2. All the species of the working fluid are gaseous. Any condensed phases
(liquid or solid) add a negligible amount to the total mass.

3. The working substance obeys the perfect gas law.

4. There is no heat transfer across the rocket walls; therefore, the flow is
adiabatic.

5. There is no appreciable friction and all boundary layer effects are
neglected.

6. There are no shock waves or discontinuities in the nozzle flow.

7. The propellant flow is steady and constant. The expansion of the working
fluid is uniform and steady, without vibration. Transient effects (i.e., start-
up and shutdown) are of very short duration and may be neglected.

8. All exhaust gases leaving the rocket have an axially directed velocity.

9. The gas velocity, pressure, temperature, and density are all uniform across
any section normal to the nozzle axis.

10. Chemical equilibrium is established within the rocket chamber and the gas
composition does not change in the nozzle (frozen flow).

11. Stored propellants are at room temperature. Cryogenic propellants are at
their boiling points.

These assumptions permit the derivation of a simple, quasi-one-dimensional
theory as mentioned in subsequent sections. Later in this book we present more
sophisticated theories or introduce correction factors for several of the items on
the list, and they allow a more accurate determination of the simplified analysis.
The next paragraph explains why these assumptions cause only small errors.

For a liquid bipropellant rocket the idealized theory postulates an injection
system in which the fuel and oxidizer are mixed perfectly so that a homogeneous
working substance results. A good rocket injector can approach this condition

3.2. SUMMARY OF THERMODYNAMIC RELATIONS 49

closely. For a solid propellant rocket unit, the propellant must essentially be
homogeneous and uniform and the burning rate must be steady. For nuclear,
solar-heated or arc-heated rockets, it is assumed that the hot gases are uniform
in temperature at any cross section and steady in flow. Because chamber tem-
peratures are typically high (2500 to 3600 K for common propellants), all gases
are well above their respective saturation conditions and actually follow the per-
fect gas law very closely. Assumptions 4, 5, and 6 above allow the use of the
isentropic expansion relations in the rocket nozzle, thereby describing the max-
imum conversion of heat to kinetic energy of the jet. This also implies that the
nozzle flow is thermodynamically reversible. Wall friction losses are difficult to
determine accurately, but they are usually negligible in nozzles. Except for very
small chambers, the energy lost as heat to the walls of the rocket is usually less
than 1% (occasionally up to 2%) of the total energy and can therefore often be
neglected. Short-term fluctuations of the steady propellant flow rate and pressure
are usually less than 5% of the rated value; their effect on rocket performance
is small and can be neglected. In well-designed supersonic nozzles, the conver-
sion of thermal energy into directed kinetic energy of the exhaust gases proceeds
smoothly and without normal shocks or discontinuities; thus the flow expansion
losses are generally small.

Some companies and some authors do not include all or the same 11 items
listed above in their definition of an ideal rocket. For example, instead of assump-
tion 8 (all nozzle exit velocity is axially directed), some use a conical exit nozzle
with a 15° half-angle as their base configuration in their ideal nozzle; this dis-
counts the divergence losses, which are described later in this chapter.

3.2. SUMMARY OF THERMODYNAMIC RELATIONS

In this section we review briefly some of the basic relationships needed for the
development of the nozzle flow equations. Rigorous derivations and discussions
of these relations can be found in many thermodynamics or fluid dynamics texts,
such as Refs. 3-1 and 3-2.

The principle of conservation of energy can be readily applied to the adiabatic,
no shaft-work process inside the nozzle. Furthermore, without shocks or friction,
the flow entropy change is zero. The concept of enthalpy is useful in flow systems;
the enthalpy comprises the internal thermal energy plus the flow work (or work
performed by the gas at a velocity v in crossing a boundary). For ideal gases
the enthalpy can conveniently be expressed as the product of the specific heat Cp
times the absolute temperature T (the specific heat at constant pressure is formally
defined as the partial derivative of the enthalpy with respect to temperature at
constant pressure). Under the above assumptions, the total or stagnation enthalpy
per unit mass ho is constant, that is,

hf) = h + jU = constant

(3-1)

50 NOZZLE THEORY AND THERMODYNAMIC RELATIONS

In the above, J is the mechanical equivalent of heat that is inserted only when
thermal units (i.e., the Btu and calorie) are mixed with mechanical units (i.e.,
the ft-lbf and the joule). In SI units (kg, m, sec) the value of J is one. In the
EE system of units the value of the constant J is given in Appendix 1. The
conservation of energy for isentropic flow between any two sections x and y
shows that the decrease in enthalpy or thermal content of the flow appears as an
increase of kinetic energy since any changes in potential energy may be neglected.

K - hy = i (u2 - vj)/J = Cp (T, -Ty) (3-2)

The principle of conservatism of mass in a steady flow with a single inlet and
single outlet is expressed by equating the mass flow rate m at any section x to
that at any other section y ; this is known in mathematical form as the continuity
equation. Written in terms of the cross-sectional area A, the velocity v, and the
specific volume V,

rhx = thy = m = AvjV (3-3)

The perfect gas law is written as

PxVx=RTx (3-4)

where the gas constant R is found from the universal gas constant R' divided
by the molecular mass 071 of the flowing gas mixture. The molecular volume
at standard conditions becomes 22.41 m^/kg-mol or ft^/lb-mol and it relates to
a value of R' = 8314.3 J/kg-mol-K or 1544 ft-lbf/lb-mol-°R. One often finds
Eq. 3-3 written in terms of density p which is the reciprocal of the specific
volume V. The specific heat at constant pressure Cp, the specific heat at constant
volume c„, and their ratio k are constant for perfect gases over a wide range of
temperatures and are related:

k = Cp/Cy (3 -5a)

Cp — Cy = R/J (3-5b)

Cp = kR/ik - l)J (3-6)

Eor an isentropic flow process the following relations hold between any points x
and y:

Tx/Ty = = iVy/Vx)^-^ (3-7)

During an isentropic nozzle expansion the pressure drops substantially, the
absolute temperature drops somewhat less, and the specific volume increases.
When a flow is stopped isentropically the prevailing conditions are known as
stagnation conditions and are designated by the subscript 0. Sometimes the word

3.2. SUMMARY OF THERMODYNAMIC RELATIONS 51

“total” is used instead of stagnation. As can be seen from Eq. 3-1 the stagnation
enthalpy consists of the sum of the static or local enthalpy and the fluid kinetic
energy. The stagnation temperature Tq is found from the energy equation as

To = r + v'^UlCpJ) (3-8)

where T is the absolute fluid static temperature. In adiabatic flows, the stagnation
temperature remains constant. The relationship of the stagnation pressure to the
local pressure in the flow can be found from the previous two equations:

po/p = [1 + = {V/Vof (3-9)

When the local velocity comes close to zero, the local temperature and pressure
will approach the stagnation pressure and stagnation temperature. In a combus-
tion chamber, where the gas velocity is small, the local combustion pressure
is essentially equal to the stagnation pressure. The velocity of sound a or the
acoustic velocity in ideal gases is independent of pressure. It is defined as

a = sfm (3-10)

In the EE system the value of R has to be corrected and the constant go is
added. Equation 3-10 becomes y/g^kRT. This correction factor must be applied
wherever R is used in EE units. The Mach number M is a dimensionless flow
parameter and is used to define the ratio of the local flow velocity v to the local
acoustic velocity a:

M = v/a = v/sf^ (3-11)

A Mach number less than one corresponds to subsonic flow and greater than
one to supersonic flow. When the Mach number is equal to one, then the flow
is moving at precisely the velocity of sound. It is shown later that at the throat
of all supersonic nozzles the Mach number must be equal to one. The relation
between stagnation temperature and Mach number can now be written from Eqs.
3-2, 3-7, and 3-10 as

TQ = T[\ + \{k-l)M^]

(3-12)

M =

Tq and pq designate the stagnation values of the temperature and pressure. Unlike
the temperature, the stagnation pressure during an adiabatic nozzle expansion
remains constant only for isentropic flows. It can be computed from

P(,=p[\ + \{k-m^t'^'~'^

(3-13)

52 NOZZLE THEORY AND THERMODYNAMIC RELATIONS

The expansion area ratio for a nozzle with isentropic flow can be expressed
in terms of Mach numbers for any points x and y within the nozzle. This rela-
tionship, along with those for the ratios T /Tq and p/po, is plotted in Fig. 3-1
for Ax = At and Mx = 1.0. Otherwise,

A;,

My^

1 -F [(^ - 1)/2]M,2 ]
1 + [(^ _ 1)/2]M,2 I

(3-14)

As can be seen from Fig. 3-1, for subsonic flow the chamber contraction ratio
AxjAf can be small, with values of 3 to 6, and the passage is convergent. There is
no noticeable effect from variations of ^ . In solid rocket motors the chamber area
A I refers to the flow passage or port cavity in the virgin grain. With supersonic
flow the nozzle section diverges and the area ratio becomes large very quickly;

' 0.10 1.0 10

Mach number

FIGURE 3-1. Relationship of area ratio, pressure ratio, and temperature ratio as func-
tions of Mach number in a De Laval nozzle for the subsonic and supersonic nozzle
regions.

3.3. ISENTROPIC FLOWTHROUGH NOZZLES 53

the area ratio is significantly influenced by the value of k. The area ratio A 2 /A,
ranges between 15 and 30 at M = 4, depending on the value of On the other
hand, pressure ratios depend little on k whereas temperature ratios show more
variation.

The average molecular mass 371 of a mixture of gases is the sum of all the
molar fractions n, multiplied by the molecular mass of each chemical species
(n,371,) and then divided by the sum of all molar mass fractions. This is further
discussed in Chapter 5. The symbol 371 is used to avoid confusion with M for the
Mach number. In many pieces of rocket literature 371 is called molecular weight.

Example 3-1. An ideal rocket is designed to operate at sea level using a propellant
whose products of combustion have a specific heat ratio k of 1.3. Determine the required
chamber pressure if the exit Mach number is 2.52. Also determine the nozzle area ratio
between the throat and exit.

SOLUTION. For optimum expansion the nozzle exit pressure must equal the local atmo-
spheric pressure, which here has a value of 0.1013 MPa. If the chamber velocity may be
neglected, then the ideal chamber pressure is the total stagnation pressure, which can be
found from Eq. 3-13 as

P0=p[1 + 3(^-1)M2]'''^<^

= 0.1013[1 -h 0.15 X 2.52^]‘-^l°-^ = 1.84 MPa

The ideal nozzle area is determined from Eq. 3-14 setting M, = 1.0 at the throat (see
also Eig. 3-1):

l + [{k- l)/2]Ml
(k + l)/2

(i-M)/2(*:-l)

2J2

1 -hO.15 X 2.52^

2.3/0.6

1.15

3.02

Note that ideal implies no losses, whereas optimum is a separate concept reflecting the
best calculated performance at a particular set of given pressures. Optimum performance
is often taken as the design condition and it occurs when p 2 = P 3 as will be shown in the
section on the thrust coefficient (the peak of the curves in Eigs. 3-7 and 3-8 for fixed
Pi/Pi)-

3.3. ISENTROPIC FLOW THROUGH NOZZLES

In a converging-diverging nozzle a large fraction of the thermal energy of
the gases in the chamber is converted into kinetic energy. As will be explained,
the gas pressure and temperature drop dramatically and the gas velocity can reach
values in excess of 2 miles per second. This is a reversible, essentially isentropic
flow process and its analysis is described here. If a nozzle inner wall has a flow
obstruction or a wall protrusion (a piece of weld splatter or slag), then the kinetic

54 NOZZLE THEORY AND THERMODYNAMIC RELATIONS

gas enery is locally converted back into thermal energy essentially equal to the
stagnation temperature and stagnation pressure in the chamber. Since this would
lead quickly to a local overheating and failure of the wall, nozzle inner walls
have to be smooth without any protrusion. Stagnation conditions can also occur
at the leading edge of a jet vane (described in Chapter 18) or at the tip of a gas
sampling tube inserted into the flow.

Velocity

From Eq. 3-2 the nozzle exit velocity i >2 can be found:

This equation applies to ideal and nonideal rockets. For constant k this expres-
sion can be rewritten with the aid of Eqs. 3-6 and 3-7. The subscripts 1 and 2
apply to the nozzle inlet and exit conditions, respectively:

This equation also holds for any two points within the nozzle. When the chamber
section is large compared to the nozzle throat section, the chamber velocity or
nozzle approach velocity is comparatively small and the term can be neglected.
The chamber temperature Tj is at the nozzle inlet and, under isentropic conditions,
differs little from the stagnation temperature or (for a chemical rocket) from the
combustion temperature. This leads to an important simplified expression of the
exhaust velocity V 2 , which is often used in the analysis:

It can be seen that the exhaust velocity of a nozzle is a function of the pressure
ratio pi/p 2 , the ratio of specific heats k, and the absolute temperature at the nozzle
inlet Ti, as well as the gas constant R. Because the gas constant for any particular
gas is inversely proportional to the molecular mass Tl, the exhaust velocity or
the specific impulse are a function of the ratio of the absolute nozzle entrance
temperature divided by the molecular mass, as is shown in Fig. 3-2. This ratio
plays an important role in optimizing the mixture ratio in chemical rockets.

(3- 15a)

(3 -15b)

(3-16)

3.3. ISENTROPIC FLOW THROUGH NOZZLES 55

I I I I 1_

50 75 100 125 150

Tl/On, K-kg-mol/kg

FIGURE 3-2. Specific impulse and exhaust velocity of an ideal rocket at optimum
nozzle expansion as functions of the absolute chamber temperature T\ and the molecular
mass Tl for several values of k and pi/p 2 -

Equations 2-14 and 2-15 give the relations between the velocity V 2 , the
thrust F, and the specific impulse Is', it is plotted in Fig. 3-2 for two pressure
ratios and three values of k. Equation 3-16 indicates that any increase in the gas
temperature (usually caused by an increase in energy release) or any decrease of
the molecular mass of the propellant (usually achieved by using light molecular
mass gases rich in hydrogen content) will improve the performance of the rocket;
that is, they will increase the specific impulse Is or the exhaust velocity V 2 or
c and, thus, the performance of the vehicle. The influences of the pressure ratio
across the nozzle p\/p 2 and of the specific heat ratio k are less pronounced. As
can be seen from Fig. 3-2, performance increases with an increase of the pressure
ratio; this ratio increases when the value of the chamber pressure p\ increases
or when the exit pressure p 2 decreases, corresponding to high altitude designs.
The small influence of k values is fortuitous because low molecular masses are
found in diatomic or monatomic gases, which have the higher values of k.

For comparing specific impulse values from one rocket system to another or
for evaluating the influence of various design parameters, the value of the pressure
ratio must be standardized. A chamber pressure of 1000 psia (6.894 Mpa) and
an exit pressure of 1 atm (0.1013 Mpa) are generally in use today.

56 NOZZLE THEORY AND THERMODYNAMIC RELATIONS

For optimum expansion p 2 — P 3 and the effective exhaust velocity c
(Eq. 2-16) and the ideal rocket exhaust velocity are related, namely

f2 = (c)opt (3-17)

and c can be substituted for V 2 in Eqs. 3-15 and 3-16. For a fixed nozzle exit
area ratio, and constant chamber pressure, this optimum condition occurs only
at a particular altitude where the ambient pressure p^ happens to be equal to the
nozzle exhaust pressure p 2 - At all other altitudes c ^ V 2 -

The maximum theoretical value of the nozzle outlet velocity is reached with
an infinite expansion (exhausting into a vacuum):

(V2)max = V^kRTolik - 1) (3-18)

This maximum theoretical exhaust velocity is finite, even though the pressure
ratio is infinite, because it represents the finite thermal energy content of the fluid.
Such an expansion does not happen, because, among other things, the temperature
of many of the working medium species will fall below their liquefaction or the
freezing points; thus they cease to be a gas and no longer contribute to the gas
expansion.

Example 3-2. A rocket operates near sea level with a chamber pressure of p\ = 2.068
MPa or 300 psia, a chamber temperature of 2222 K, and a propellant consumption of
m = 1.0 kg/sec. Take k = 1.30 and R = 345.7 J/kg-K. Calculate the ideal thmst and the
ideal specific impulse. Also plot the cross-section area A, the local velocity v, the specific
volume V, the absolute temperature T, and the local Mach number M with respect to
pressure along the nozzle.

SOLUTION. Assume that the operation of this rocket is at an optimum for expansion to
sea-level pressure p 3 = 0.1013 MPa so thal the ratio pi/pi = 0.049 in Eq. 3-16. From
this equation, the effective exhaust velocity is calculated to be 1827 m/sec since here
V 2 = c. Hence,

I = c/go = 1827/9.81 = 186 sec and
F = mc = 1.0 X 1827 = 1827 N

In order to graph the desired variables as a function of pressure along the nozzle, we
have to express them in analytically useful ways. In this chapter, we denote the noz-
zle axial location x as either the chamber (location 1) or the throat (location t) and keep
the axial location y as a variable which spans the range 1^2 (see Fig. 2-1). The initial
specific volume Vi is calculated from the equation of state of a perfect gas (Eq. 3-4),

Pi = RTi/pi = 345.7 X 2222/(2.068 x 10®) = 0.3714 m^kg

Now we can write equations for the specific volume and the temperature (Eq. 3-7) as

V, = Pi(pi/p,)'/*^ =0.3714(2.068/p,)°’®® (m^/kg)

Ty = ri(p,/pi)<*-‘>/*^ = 2222(py/2.06Sf^^^ (K)

3.3. ISENTROPIC FLOW THROUGH NOZZLES 57

These can be plotted by inserting py values between 2.068 and 0.1013 in MPa. The
calculations for the velocity follow from Eq. 3-16 (the derivations have been omitted in
order to focus on the plotting and are left up to the interested reader),

Vy = 2580[1 - (Pj,/ 2.068)“-^^']‘''2 (m/sec)

The cross-sectional area is found from Eq. 3-3 and the relations above:

Aj, = mVy/Vy = (1.0 X 0.3714/2580) x (2.068/p3,)°™

X [1 - (Pj,/2.068)“-^^']-'/2

= 1.44 X lO-'^ X (2.068 /p 3,)°-’®[1 - (py/2.068)°-^^']“'''^ (m^)
Einally the Mach number is obtained using Eq. 3-11 and the relations above:

My = Vy/[kRTy]^''^ = [2580/(1.30 x 345.7 x 2222)'/^]

X [1 - (Pj,/2.068)“^^]'/^(p,/2.068)“°-2^‘

= 2.582[1 - (p3,/2.068)°-^^‘]'/^(Pj,/2.068)“°-^^‘

With the equations indicated above and graphing software such as available in MATLAB
or MAPLE, the desired plots may be obtained. Figure 3-3 shows semiquantitative plots
of several parameter variations and should be used for comparison. Note that in this figure
the pressures are shown as decreasing to the right, representing a flow from left to right
(also its units are in psia where the throat pressure is 164.4 psia and the exit pressure
would be 14.7 psia).

A number of interesting deductions can be made from this example. Very high
gas velocities (over 1 km/sec) can be obtained in rocket nozzles. The temperature
drop of the combustion gases flowing through a rocket nozzle is appreciable. In
the example given the temperature changed 1 1 15°C in a relatively short distance.
This should not be surprising, for the increase in the kinetic energy of the gases is
derived from a decrease of the enthalpy, which in turn is roughly proportional to
the decrease in temperature. Because the exhaust gases are still very hot (1107 K)
when leaving the nozzle, they contain considerable thermal energy not available
for conversion into kinetic energy of the jet.

Nozzle Flow and Throat Condition

The required nozzle area decreases to a minimum (at 1.130 MPa or 164 psi pres-
sure in the previous example) and then increases again. Nozzles of this type (often
called De Laval nozzles after their inventor) consist of a convergent section fol-
lowed by a divergent section. From the continuity equation, the area is inversely
proportional to the ratio v/V . This quantity has also been plotted in Fig. 3-3.
There is a maximum in the curve of v/V because at first the velocity increases

58 NOZZLE THEORY AND THERMODYNAMIC RELATIONS

Pressure, mPa

Nozzle inlet

FIGURE 3-3. Typical variation of cross-sectional area, temperature, specific volume,
and velocity with pressure in a rocket nozzle (from Example 3-2).

at a greater rate than the specific volume; however, in the divergent section, the
specific volume increases at a greater rate.

The minimum nozzle area is called the throat area. The ratio of the nozzle
exit area to the nozzle throat area A, is called the nozzle expansion area ratio
and is designated by the Greek letter e. It is an important nozzle parameter:

c = Ml At

(3-19)

3.3. ISENTROPIC FLOWTHROUGH NOZZLES 59

The maximum gas flow per unit area occurs at the throat where there is a
unique gas pressure ratio which is only a function of the ratio of specific heats
k. This pressure ratio is found by setting M = 1 in Eq. 3-13:

The throat pressure pt for which the isentropic mass flow rate is a maximum
is called the critical pressure. Typical values of this critical pressure ratio range
between 0.53 and 0.57 of the inlet pressure. The flow through a specified rocket
nozzle with a given inlet condition is less than the maximum if the pressure
ratio is larger than that given by Eq. 3-20. However, note that this ratio is not
that across the entire nozzle and that the maximum flow or choking condition
(explained below) is always established internally at the throat and not at the
exit plane. The nozzle inlet pressure is very close to the chamber stagnation
pressure, except in narrow combustion chambers where there is an appreciable
drop in pressure from the injector region to the nozzle entrance region. This is
discussed in Section 3.5. At the point of critical pressure, namely the throat, the
Mach number is one and the values of the specific volume and temperature can
be obtained from Eqs. 3-7 and 3-12:

In Eq. 3-22 the nozzle inlet temperature Ti is very close to the combustion
temperature and hence close to the nozzle flow stagnation temperature Tq. At the
critical point there is only a mild change of these properties. Take, for example,
a gas with k = 1.2; the critical pressure ratio is about 0.56 (which means that p,
equals almost half of the chamber pressure pi ; the temperature drops only slightly
{Tt = 0.91Ti), and the specific volume expands by over 60% {Vt = 1.61Vi).
From Eqs. 3-15, 3-20, and 3-22, the critical or throat velocity u, is obtained:

The first version of this equation permits the throat velocity to be calculated
directly from the nozzle inlet conditions without any of the throat conditions
being known. At the nozzle throat the critical velocity is clearly also the sonic
velocity. The flow in the chamber is subsonic and downstream of the nozzle throat
it is supersonic. The divergent portion of the nozzle permits further decreases in
pressure and increases in velocity under supersonic conditions. If the nozzle is cut
off at the throat section, the exit gas velocity is sonic and the flow rate remains
a maximum. The sonic and supersonic flow condition can be attained only if the
critical pressure prevails at the throat, that is, if pi/pi is equal to or less than
the quantity defined by Eq. 3-20. There are, therefore, three different types of
nozzles: subsonic, sonic, and supersonic, and these are described in Table 3-1.

P,!p, = [2/{k + l)]'=/(^-D

(3-20)

V, = Vi[(k + l)/2]'/(^-^)
T, = 2Ti/(k + 1)

(3-21)

(3-22)

(3-23)

60 NOZZLE THEORY AND THERMODYNAMIC RELATIONS

TABLE 3-1. Nozzle Types

Subsonic Sonic Supersonic

Throat

velocity

Exit

velocity

Mach

number

Pressure

ratio

Shape

vi < at Vt = at Vt = at

V2 < a2 V2 = Vt V2 > Vt

M 2 <1 M2=M, = 1.0 M 2 > 1

The supersonic nozzle is the one used for rockets. It achieves a high degree
of conversion of enthalpy to kinetic energy. The ratio between the inlet and exit
pressures in all rockets is sufficiently large to induce supersonic flow. Only if
the absolute chamber pressure drops below approximately 1.78 atm will there be
subsonic flow in the divergent portion of the nozzle during sea-level operation.
This condition occurs for a very short time during the start and stop transients.

The velocity of sound is equal to the propagation speed of an elastic pres-
sure wave within the medium, sound being an infinitesimal pressure wave. If,
therefore, sonic velocity is reached at any point within a steady flow system, it is
impossible for a pressure disturbance to travel past the location of sonic or super-
sonic flow. Thus, any partial obstruction or disturbance of the flow downstream
of the nozzle throat with sonic flow has no influence on the throat or upstream
of it, provided that the disturbance does not raise the downstream pressure above
its critical value. It is not possible to increase the throat velocity or the flow
rate in the nozzle by further lowering the exit pressure or even evacuating the
exhaust section. This important condition is often described as choking the flow.
It is always established at the throat and not the nozzle exit plane. Choked flow
through the critical section of a supersonic nozzle may be derived from Eqs.
3-3, 3-21, and 3-23. It is equal to the mass flow at any section within the
nozzle.

m =

AtVt

= AtPik

y/[2/{k + l)]«=+l)/dr-l)

VMtY

(3-24)

The mass flow through a rocket nozzle is therefore proportional to the throat
area A, and the chamber (stagnation) pressure pp, it is also inversely proportional
to the square root of T /9J1 and a function of the gas properties. For a supersonic
nozzle the ratio between the throat and any downstream area at which a pressure

3.3. ISENTROPIC FLOWTHROUGH NOZZLES 61

Py prevails can be expressed as a function of the pressure ratio and the ratio of
specific heats, by using Eqs. 3-4, 3-16, 3-21, and 3-23, as follows:

^ _ VtVy
Ay VyVt

1/L

k + \
k - 1

1 - 1 ^
Pi

(k-iyk-

(3-25)

When Py = p 2 , then Ay j A, = A 2 IA, = e in Eq. 3-25. Eor low-altitude operation
(sea level to about 10,000 m) the nozzle area ratios are typically between 3 and 25,
depending on chamber pressure, propellant combinations, and vehicle envelope
constraints. Eor high-altitude (100 km or higher) area ratios are typically between
40 and 200, but there have been some as high as 400. Similarly, an expression for
the ratio of the velocity at any point downstream of the throat with the pressure
Py, and the throat velocity may be written from Eqs. 3-15 and 3-23:

Vy _

k + \

k - 1

\pi/

(3-26)

These equations permit the direct determination of the velocity ratio or the
area ratio for any given pressure ratio, and vice versa, in ideal rocket nozzles.
They are plotted in Eigs. 3-4 and 3-5, and these plots allow the determination
of the pressure ratios given the area or velocity ratios. When py = p 2 , Eq. 3-26
describes the velocity ratio between the nozzle exit area and the throat section.
When the exit pressure coincides with the atmospheric pressure (p 2 = P 3 , see
Eig. 2-1), these equations apply for optimum nozzle expansion. Eor rockets
that operate at high altitudes, not too much additional exhaust velocity can be
gained by increasing the area ratio above 1000. In addition, design difficulties
and a heavy inert nozzle mass make applications above area ratios of about 350
marginal.

Appendix 2 is a table of several properties of the earth’s atmosphere with
agreed-upon standard values. It gives ambient pressure for different altitudes.
These properties can vary somewhat from day to day (primarily because of solar
activity) and between hemispheres. Eor example, the density of the atmosphere
at altitudes between 200 and 3000 km can change by more than an order of
magnitude, affecting satellite drag.

Example 3-3. Design an ideal nozzle for a rocket that operates at 25 km altitude and
delivers 5000 N thrust with a chamber pressure of 2.039 MPa and a chamber temperature
of 2800 K. Assuming that k = \.2Q and R = 360 J/kg-K, determine the throat area, exit
area, throat temperature, and exit velocity.

SOLUTION. At 25 km altitude the atmospheric pressure equals 2.549 kPa (in Appendix
2 the ratio is 0.025158, which must be multiplied by the pressure at sea level, 0.1013
MPa). The controlling pressure ratio is

P 2 /P 1 = 0.002549/2.039 = 0.00125 = 1/800

62 NOZZLE THEORY AND THERMODYNAMIC RELATIONS

10 15 20 25 30 40 50 60 80 100 150 200 300

FIGURE 3-4. Area and velocity ratios as function of pressure ratio for the diverging
section of a supersonic nozzle.

The area ratio can now he obtained directly from Eq. 3-25 (or Fig. 3-5) as

A, a + U + i

Ai \ 2 } U/ ^k-l

= (1.10)^(0.00125)°*^Vll-0[l - (0.00125)‘’ 'S’]

= 1.67 X 10“^ = 1/60

(k-i)/k

The temperature at the throat may be found directly from Eq. 3-22:
T, = 2Ti/(k + 1) = 2 X 2800/2.2 = 2545 K
The ideal exit velocity is found from Eq. 3-15 as

k - 1

RTi

1-1^
Pi

yi.21 X 10’[1 - (0.00125)01®’]

= 2851 m/sec

3.3. ISENTROPIC FLOWTHROUGH NOZZLES 63

300 500 1000 2000 3000 5000 10,000

FIGURE 3-5. Continuation of prior figure of area ratios and velocity ratios, but for
higher pressure ratios in a supersonic nozzle.

Next we need to find the throat area, which can be found via the mass flow rate, Eq. 3-24,
A, = mVt/vt. To find the mass flow rate explicitly we note that V 2 = c (Eq. 2-16) and
from Eq. 2-17

m = Fjc = 5000/2851 = 1.754 kg/sec
A = ” /

' Pi Y /t[2/(fe + l)](*+i)/(^-0

1.754 / 360 X 2800

2.039 X lO^Y 1.2[2/2.2]ii

= 13.32 cm^

And thus the exit area becomes A 2 = 60 x 13.32 = 799 cm^. The designer would next
have to modify the area ratio of the nozzle depending on the actual configuration cho-
sen and to choose materials together with cooling methods that accommodate the high
chamber and throat temperatures.

64 NOZZLE THEORY AND THERMODYNAMIC RELATIONS

Thrust and Thrust Coefficient

The efflux of the propellant gases or the momentum flowing out causes the
thrust or reaction force on the rocket structure. Because the flow is supersonic,
the pressure at the exit plane of the nozzle may he different from the ambient
pressure, and the pressure thrust component adds to the momentum thrust as
given by Eq. 2-14, which is repeated here:

F = mV2 + (P2 -P3)M

The maximum thrust for any given nozzle operation is found in a vacuum
where = 0. Between sea level and the vacuum of space, Eq. 2-14 gives the
variation of thrust with altitude, using the properties of the atmosphere such as
those listed in Appendix 2. Eigure 2-2 shows a typical variation of thrust with
altitude. To modify values calculated for optimum operating conditions (p 2 = Pi)
for given values of pi, k, and A 2 /A,, the following expressions may be used. Eor
the thrust.

F = Fopi+piAf

Pi) A,

For the specific impulse, using Eqs. 2-5, 2-18, and 2-14,

c € ( P 2

Is = (/Jopt + -

go \Pl

(3-27)

(3-28)

If, for example, the specific impulse for a new exit pressure p 2 corresponding to
a new area ratio A 2 /A, is to be calculated, the above relations may be used.

Equation 2-14 can be expanded by modifying it and substituting V 2 ,v,, and
Vt from Eqs. 3-16, 3-21, and 3-23:

A,V,V2

F = — h (P2 -Pi)A2

= A,pi

2k^

k — 1 V A: + 1

(k+i)/(k-i)

(k-D/k-

+ (P2- Pi) A 2

(3-29)

The first version of this equation is general and applies to all rockets, the sec-
ond form applies to an ideal rocket with k being constant throughout the expan-
sion process. This equation shows that the thrust is proportional to the throat area
At and the chamber pressure (or the nozzle inlet pressure) pi and is a function
of the pressure ratio across the nozzle pi/p 2 , the specific heat ratio k, and of the
pressure thrust. It is called the ideal thrust equation. The thrust coefficient Cf is

3.3. ISENTROPIC FLOWTHROUGH NOZZLES 65

defined as the thrust divided by the chamber pressure p\ and the throat area A,.
Equations 2-14, 3-21, and 3-16 then give this relation:

^ U2A2 ^ P2A2 P3A2
PiA,V2 Pi At piA,

2k^

k-l\k+l

(k+\)/(k-i)

i-r-

(k-i)/k-

+ — — —— (3-30)
Pi At

The thrust coefficient Cp is dimensionless. It is a key parameter for analysis
and is a function of gas property k, the nozzle area ratio e, and the pressure
ratio across the nozzle pi/pi, but independent of chamber temperature. For any
hxed pressure ratio pijps, the thrust coefficient Cp and the thrust F have a peak
when p 2 = Pi- This peak value is known as the optimum thrust coefficient and
is an important criterion in nozzle design considerations. The use of the thrust
coefficient permits a simplihcation to Eq. 3-29:

F = CpAtpi (3-31)

Equation 3-31 can be solved for Cp and provides the relation for determining
the thrust coefficient experimentally from measured values of chamber pres-
sure, throat diameter, and thrust. Even though the thrust coefficient is a function
of chamber pressure, it is not simply proportional to pi, as can be seen from
Eq. 3-30. However, it is directly proportional to throat area. The thrust coef-
ficient can be thought of as representing the amplification of thrust due to the
gas expanding in the supersonic nozzle as compared to the thrust that would
be exerted if the chamber pressure acted over the throat area only. The thrust
coefficient has values ranging from about 0.8 to 1.9. It is a convenient parameter
for seeing the effects of chamber pressure or altitude variations in a given nozzle
configuration, or to correct sea-level results for flight altitude conditions.

Figure 3-6 shows the variation of the optimum expansion thrust coefficient
(p 2 = pf) for different pressure ratios p\lp 2 , values of k, and area ratio e. The
complete thrust coefficient is plotted in Figs. 3-7 and 3-8 as a function of
pressure ratio p\/p 2 and area ratio for k = 1.20 and 1.30. These two sets of curves
are useful in solving various nozzle problems for they permit the evaluation of
under- and over-expanded nozzle operation, as explained below. The values given
in these figures are ideal and do not consider such losses as divergence, friction,
or internal expansion waves.

When p\fp 2 , becomes very large (e.g., expansion into near vacuum), then the
thrust coefficient approaches an asymptotic maximum as shown in Figs. 3-7 and
3-8. These hgures also give values of for any mismatched nozzle (p 2 Ps),
provided the nozzle is flowing full at all times, that is, the working fluid does
not separate or break away from the walls. Flow separation is discussed later in
this section.

FIGURE 3-6. Thrust coefficient Cp as a function of pressure ratio, nozzle area ratio, and specific heat ratio for optimum expansion conditions

3.3. ISENTROPIC FLOWTHROUGH NOZZLES 67

Area ratio e = Ag/A,

FIGURE 3-7. Thrust coefficient Cf versus nozzle area ratio for k = 1.20.

FIGURE 3-8. Thrust coefficient Cf versus nozzle area ratio for k = 1.30.

68 NOZZLE THEORY AND THERMODYNAMIC RELATIONS

Characteristic Velocity and Specific Impulse

The characteristic velocity c* was defined by Eq. 2-18. From Eqs. 3-24 and
3-31 it can be shown that

^ ^ ^ JL = ^/kRTl

^ m Cf Cf ky/[2l(k + l)](*:+l)/d^-l)

It is basically a function of the propellant characteristics and combustion chamber
design; it is independent of nozzle characteristics. Thus, it can be used as a figure
of merit in comparing propellant combinations and combustion chamber designs.
The first version of this equation is general and allows the determination of c*
from experimental data of m,pi, and A,. The last version gives the maximum
value of c* as a function of gas properties, namely k, the chamber temperature,
and the molecular mass 971, as determined from the theory in Chapter 5. Some
values of c* are shown in Tables 5-4 and 5-5.

The term c*-efficiency is sometimes used to express the degree of completion
of the energy release and the creation of high-temperature, high-pressure gas
in the chamber. It is the ratio of the actual value of c*, as determined from
measurements (the first part of Eq. 3-32), and the theoretical value (last part of
Eq. 3-32), and typically has a value between 92 and 99.5%.

Using Eqs. 3-31 and 3-32, the thrust itself may now be expressed as the
mass flow rate times a function of the combustion chamber (c*) times a function
of the nozzle expansion Cf),

F = CFihc* (3-33)

Some authors use a term called the discharge coefficient Co , which is merely
the reciprocal of c*. Both Co and the characteristic exhaust velocity c* are used
primarily with chemical rocket propulsion systems.

The influence of variations in the specific heat ratio k on various parame-
ters (such as c,c*A 2 lAt,V 2 lvu or f) is not as large as the changes in chamber
temperature, pressure ratio, or molecular mass. Nevertheless, it is a noticeable
factor, as can be seen by examining Figs. 3-2 and 3-4 to 3-8. The value of k is
1.67 for monatomic gases such as helium and argon, 1.4 for cold diatomic gases
such as hydrogen, oxygen, and nitrogen, and for triatomic and beyond it varies
between 1.1 and 1.3 (methane is 1.11 and ammonia and carbon dioxide 1.33).
In general, the more complex the molecule the lower the value of k\ this is also
true for molecules at high temperatures when their vibrational modes have been
activated. The average values of k and 971 for typical rocket exhaust gases with
several constituents depend strongly on the composition of the products of com-
bustion (chemical constituents and concentrations), as explained in Chapter 5.
Values of k and 971 are given in Tables 5-4, 5-5, and 5-6.

Example 3-4. What is the percentage variation in thrust between sea level and 10 km
for a launch vehicle whose rocket operates with a chamber pressure of 20 atm and has

3.3. ISENTROPIC FLOWTHROUGH NOZZLES 69

an expansion area ratio of 6? (Use k = 1.30.) Also, at what altitude is the performance
of this rocket optimum?

SOLUTION. According to Eq. 3-33, the only component of the thrust that depends
on the ambient pressure is Cp, the thrust coefficient. This coefficient can be found from
Eq. 3-30 or from Fig. 3-8 by following a vertical line corresponding to AijAt = 6.0. We
first have to find the ratio pi/ps at sea level and at 10 km. Using values from Appendix 2
and solving Eq. 3-25 for P 2 /P 3 = 0.0198(p2 = 0.396 atm).

Sea level: pi/p 3 = 20/1.0 = 20 Cf = 1.33

10 km altitude: pi/p 3 = 20/0.26151 = 76.5 Cp = 1.56

Thrust increase = (1.56 - 1.33)/1.33 = 17.3%

Note that performance at optimum nozzle expansion (p 2 = P 3 ) this nozzle delivers a
thrust that is somewhat less than the maximum. From Fig. 3-6 we find that for e = 6
the optimum Cp = 1.52, which corresponds to 7.2 km elevation, whereas the maximum
value is Cp = 1.63 in a vacuum. The thrust coefficient could be somewhat higher for
larger area ratios.

Under- and Over-Expanded Nozzles

An under-expanded nozzle discharges the fluid at an exit pressure greater than
the external pressure because the exit area is too small for an optimum area ratio.
The expansion of the fluid is therefore incomplete within the nozzle, and further
expansion will take place outside of the nozzle exit. The nozzle exit pressure is
higher than the local atmospheric pressure.

In an over-expanded nozzle the fluid exists at lower pressure than the
atmosphere as it has an exit area too large for optimum. The phenomenon of
over-expansion for a supersonic nozzle is shown in Fig. 3-9, with typical
pressure measurements of superheated steam along the nozzle axis and different
back pressures or pressure ratios. Curve AB shows the variation of pressure
with the optimum back pressure corresponding to the nozzle area ratio. Curves
AC and AD show the variation of pressure along the axis for increasingly
higher external pressures. The expansion within the nozzle proceeds normally
for the initial portion of the nozzle. At point I on curve AD, for example, the
pressure is lower than the exit pressure and a sudden rise in pressure takes place
that is accompanied by the separation of the flow from the walls (separation is
described later).

The nonideal behavior of nozzles is strongly influenced by the presence of
compression waves or shock waves inside the diverging nozzle section, which are
strong compression discontinuities and exist only in supersonic flow. The sudden
pressure rise in the curve ID is such a compression wave. Expansion waves, also
strictly supersonic phenomena, match the flow from a nozzle exit to lower ambi-
ent pressures. Compression and expansion waves are described in Chapter 20.

The different possible flow conditions in a supersonic nozzle are as follows:

1. When the external pressure p 3 is below the nozzle exit pressure p 2 , the
nozzle will flow full but will have external expansion waves at its exit (i.e.,

70 NOZZLE THEORY AND THERMODYNAMIC RELATIONS

Divergence

FIGURE 3-9. Distribution of pressures in a converging-diverging nozzle for different
flow conditions. Inlet pressure is the same, but exit pressure changes. Based on experi-
mental data from A. Stodala.

under-expansion). The expansion of the gas inside the nozzle is incomplete
and the values of Cf and 4 will be less than at optimum expansion.

2. For external pressures slightly higher than the nozzle exit pressure p 2 ,
the nozzle will continue to flow full. This occurs until p 2 reaches a value
between about 10 and 40% of p^. The expansion is somewhat inefficient
and Cf and 4 will have lower values than an optimum nozzle would have.
Shock waves will exist outside the nozzle exit section.

3. For higher external pressures, separation of the flow will begin to take place
inside the divergent portion of the nozzle. The diameter of the supersonic jet
will be smaller than the nozzle exit diameter. With steady flow, separation
is typically axially symmetric. Figures 3-10 and 3-11 show diagrams of
separated flows. The axial location of the separation plane depends on
the local pressure and the wall contour. The point of separation travels
downstream with decreasing external pressure. At the nozzle exit the flow
in the center portion remains supersonic but is surrounded by an annular-
shaped section of subsonic flow. There is a discontinuity at the separation
location and the thrust is reduced, compared to a nozzle that would have

3.3. ISENTROPIC FLOW THROUGH NOZZLES 71

Altitude, m

FIGURE 3-10. Thrust coefficient Cp for two nozzles with different area ratios. One has
jet separation below about 7000 m altitude. The fully expanded exhaust plume at higher
altitudes is not shown in the sketch.

been cut off at the separation plane. Shock waves exist outside the nozzle
in the external plume.

4. For nozzles in which the exit pressure is just below the value of the inlet
pressure, the pressure ratio is below the critical pressure ratio (as defined
by Eq. 3-20) and subsonic flow prevails throughout the entire nozzle. This
condition occurs normally in rocket nozzles for a short time during the start
and stop transients.

The method for estimating pressure at the location of the separation plane
inside the diverging section of a supersonic nozzle has usually been empirical.
Reference 3-4 describes a variety of nozzles, their behavior, and methods used
to estimate the location and the pressure at separation. Actual values of pressure
for the over-expanded and under-expanded regimes described above are functions
of the specific heat ratio and the area ratio (see Ref. 3-1).

The axial thrust direction is not usually altered by separation because a steady
flow usually separates uniformly over a cross section in a divergent nozzle cone
of conventional rocket design. During transients, such as start and stop, the sep-
aration may not be axially symmetric and may cause momentary but large side
forces on the nozzle. During a normal sea-level transient of a large rocket nozzle

72 NOZZLE THEORY AND THERMODYNAMIC RELATIONS

During flight During sea level static tests

FIGURE 3-11. Simplified sketches of exhaust gas behavior of three typical rocket noz-
zles for a three-stage launch vehicle. The first vehicle stage has the biggest chamber and
the highest thrust but the lowest nozzle area ratio, and the top or third stage usually has
the lower thrust but the highest nozzle area ratio.

(before the chamber pressure reaches its full value), some momentary flow oscil-
lations and nonsymmetric separation of the jet can occur during over-expanded
flow operation. The magnitude and direction of transient side forces can change
rapidly and erratically. The resulting side forces can be large and have caused
failures of nozzle exit cone structures and thrust vector control gimbal actua-
tors. References 3-4, 3-5, and 3-6 discuss techniques for estimating these side
forces.

When the flow separates, as it does in a highly over-expanded nozzle, the
thrust coefficient Cp can be estimated if the point of separation in the nozzle is
known. Thus, Cf can be determined for an equivalent smaller nozzle with an
exit area equal to that at the point of separation. The effect of separation is to
increase the thrust and the thrust coefficient over the value that they would have
if separation had not occurred. Thus, with separated gas flow, a nozzle designed
for high altitude (large value of e) would have a larger thrust at sea level than
expected, but not as good as an optimum nozzle; in this case separation may
actually be desirable. With separated flow a large and usually heavy portion of
the nozzle is not utilized and the nozzle is bulkier and longer than necessary. The
added engine weight and size decrease flight performance. Designers therefore
select an area ratio that will not cause separation.

3.3. ISENTROPIC FLOWTHROUGH NOZZLES 73

Because of uneven flow separation and potentially destructive side loads, sea-
level static tests of an upper stage or a space propulsion system with a high area
ratio over-expanded nozzle are usually avoided; instead, a sea-level test nozzle
with a much smaller area ratio is substituted. However, actual and simulated alti-
tude testing (in an altitude test facility similar to the one described in Chapter 21)
would be done with a nozzle having the correct large area ratio. The ideal solu-
tion that avoids separation at low altitudes and has high values of Cp at high
altitudes is a nozzle that changes area ratio in flight. This is discussed at the end
of this section.

For most applications, the rocket system has to operate over a range of alti-
tudes; for a fixed chamber pressure this implies a range of nozzle pressure ratios.
The condition of optimum expansion (p 2 = ps) occurs only at one altitude, and
a nozzle with a fixed area ratio is therefore operating much of the time at either
over-expanded or under-expanded conditions. The best nozzle for such an appli-
cation is not necessarily one that gives optimum nozzle gas expansion, but one
that gives the largest vehicle flight performance (say, total impulse, or specific
impulse, or range, or payload); it can often be related to a time average over the
powered flight trajectory.

Example 3-5. Use the data from Example 3-4 (pi = 20 atm, fc = 1.30) and consider
both the area ratio 6 = 6.0 and e = 15.0. Compare the performance of these two nozzles
as a function of altitude by plotting Cp up to 50 km. Assume no shocks inside the nozzle.

SOLUTION. The procedures are the same as those in Example 3-4 and the resulting
plot is shown in Fig. 3-10. The values of Cp can be obtained by following vertical lines
for e = 6.0 and 15.0 in Fig. 3-8 corresponding to different pressure ratios at increasing
altitudes. Alternatively, Eq. 3-30 may be used for greater accuracy. The lower area ratio
gives a better performance at the lower altitudes but above about 10 km the larger area
ratio gives a higher Cp. Also noted in Fig. 3-10 by the dotted line is what a continuously
variable nozzle area ratio could accomplish by matching p 2 to pa at all altitudes.

The optimum pressure ratio always occurs when pi/pa = p\lpi and for the e = 15.0
case, from Fig. 3-6 or 3-8, this value is about 180; pa = 20/180 = 0.111 atm, which
corresponds to about 15.5 km altitude. Below this altitude the nozzle is over-expanded.
At sea level, where pi/pa =20, separation could occur as indicated in Fig. 3-8. For
similarly shaped nozzles, it is estimated that sea-level separation begins to take place at
a cross section where the local pressure is between 10 and 40% of pa or below 0.4 atm.
Upon separation, this nozzle would not flow full downstream of an area ratio of about 6
or 7 and the gas jet would occupy only the central portion of the nozzle exit. See Figs.
3-10 and 3-11. Weak shock waves and jet contraction would then raise the exhaust jet’s
pressure to match the value of the surrounding atmosphere. If the jet does not separate,
it would reach an exit pressure of 0. 1 1 atm, but this is often an unstable condition. As
the vehicle gains altitude, the separation plane would gradually move downstream until,
at an altitude of about 700 m, the exhaust gases would occupy the full diverging section
of the nozzle area.

The values of Cp can be obtained by following a vertical line for e = 15 and e = 6
in Fig. 3-8 for different pressure ratios, which correspond to different altitudes. Alterna-
tively, Eq. 3-30 can be used for better accuracy. Results are similar to those plotted in
Fig. 3-10. The lower area ratio of 6 gives a higher Cp at low altitudes but is inferior to

74 NOZZLE THEORY AND THERMODYNAMIC RELATIONS

the area ratio of 15 at the higher altitudes. A nozzle area ratio larger than 15 would yield
a higher Cp at the higher altitudes.

Figure 3-11 shows a comparison of altitude and sea-level behavior of three
nozzles and their plumes at different area ratios for a typical three-stage satellite
launch vehicle. When fired at sea-level conditions, the nozzle of the third stage
with the highest area ratio will experience flow separation and suffer a major
performance loss; the second stage will flow full but the external plume will
contract; since p 2 < Pi, there is a loss in /j and F . There is no effect on the first
stage nozzle.

Example 3-6. A rocket engine test near sea level gives the following data: thrust F =
53,000 Ibf, propellant mass flow rate m = 208 Ibm/sec, nozzle exit area ratio A^jAt =
10.0, actual local atmospheric pressure at test station = 13.8 psia, and chamber pressure
p\ = 620 psia. The test engineer also knows that for the same flow rate (and mixture
ratio) the theoretical specific impulse is 289 sec at the standard reference conditions of
Pi = 1000 psia and p 3 = 14.7 psia, and that k = 1.20. For many propellants we may
assume that the combustion temperature and k do not vary significantly with chamber
pressure. Compare the test performance of this rocket with its equivalent at sea level,
standard, and vacuum performance.

SOLUTION. The pressure ratio for the test condition is 620/13.8 = 44.93; if the test
would have been conducted at 14.7 psia, this pressure ratio would have been 620/14.7 =
42.18; for the standard reference conditions the pressure ratio is 1000/14.7 = 68.03.

Since this nozzle is operating supersonically, we find, from Eqs. 2-5, 3-24, and
3-31, that specific impulse ratios depend only on ratios of the thrust coefficient, that is,
(Di, = (Is)al{CF)h / (CF)a]- From Fig. 3-7 or from Eq. 3-30 one obtains Cp = 1.48 for
the test condition with e = 10,k = 1.20, and pi/ps = 44.9. The corresponding specific
impulse is, from Eq. 2-5, L = 53000/208 = 255 sec.

The following table shows all the results:

Test

Results

Test for
Sea Level

Standard

Condition

Vacuum

Condition

Pi (psia)

620

1,000

620

P 3 (psia)

13.8

14.7

Pl/P3

44.93

42.18

68.03

F (Ibf)

53,000

52,494

55,744

60,741

h (sec)

255

252

268

292

1.52

1.50

1.60

1.74

Figures 3-10 and 3-11 suggest that an ideal design for an ascending (e.g.,
launch) rocket vehicle would have a “rubber-like” diverging section that could be
lengthened and enlarged so that the nozzle exit area could be made larger as the
ambient pressure is reduced. The design would then allow the rocket vehicle to
attain its maximum performance at all altitudes as it ascends. As yet we have not
achieved a simple mechanical hardware design with this full altitude compen-
sation similar to “stretching rubber.” However, there are a number of practical

3.4. NOZZLE CONFIGURATIONS 75

TABLE 3-2. Estimated Losses for Small-Diameter Chambers

Chamber-to-Throat
Area Ratio

Throat
Pressure (%)

Thrust

Reduction (%)

Specific Impulse
Reduction (%)

100

3.5

1.5

0.31

2.0

5.0

0.55

1.0

19.5

1.34

k = 1.20; Pi /p2 = 1000.

nozzle configurations that can be used to alter the flow shape with altitude and
obtain maximum performance at different altitudes. They are discussed in the
next section.

Influence of Chamber Geometry

When the chamber has a cross section that is larger than about four times the
throat area (Ai/A, > 4), the chamber velocity ui, can be neglected, as was men-
tioned in explaining Eqs. 3-15 and 3-16. However, vehicle space or weight
constraints often require smaller thrust chamber areas for liquid propellant engines
and grain design considerations lead to small void volumes or small perforations
or port areas for solid propellant motors. Then iq can no longer be neglected as
a contribution to the performance. The gases in the chamber expand as heat is
being added. The energy necessary to accelerate these expanding gases within
the chamber will also cause a pressure drop and an additional energy loss. This
acceleration process in the chamber is adiabatic (no heat transfer) but not isen-
tropic. This loss is a maximum when the chamber diameter is equal to the nozzle
diameter, which means that there is no converging nozzle section. This has been
called a throatless rocket motor and has been used in a few tactical missile booster
applications, where there was a premium on minimum inert mass and length. The
flight performance improvement due to inert mass savings supposedly outweighs
the nozzle performance loss of a throatless motor. Table 3-2 lists some of the
performance penalties for three chamber area ratios.

Because of this pressure drop within narrow chambers, the chamber pressure
is lower at the nozzle entrance than it would be if Ai/A, had been larger. This
causes a small loss in thrust and specific impulse. The theory of this loss is given
in Ref. 3-7 and some results are listed in Table 3-2.

3.4. NOZZLE CONFIGURATIONS

A number of different proven nozzle configurations are available today. This
section describes their geometries and performance. Other chapters (6, 8, 12,
15, and 18) discuss their materials, heat transfer, or application and mention

76 NOZZLE THEORY AND THERMODYNAMIC RELATIONS

their requirements, design, construction, and thrust vector control. Nozzles and
chambers are usually of circular cross section and have a converging section, a
throat at the narrowest location (minimum cross section), and a diverging section.
Nozzles can be seen in Figs. 1-4, 1-5, 1-8, 2-1, 3-11, 3-12, 3-14, 11-1,
12-1 to 12-3, and 15-6 to 15-8. Refs. 3-4 and 3-8 describe many nozzle
configurations.

The converging nozzle section between the chamber and the nozzle throat
has never been critical in achieving high performance. The subsonic flow in this
section can easily be turned at very low pressure drop and any radius, cone angle,
wall contour curve, or nozzle inlet shape is usually satisfactory. A few small
attitude control thrust chambers have had their nozzle at 90° from the combustion
chamber axis without any performance loss. The throat contour also is not very
critical to performance, and any radius or other curve is usually acceptable. The
pressure gradients are high in these two regions and the flow will adhere to the
walls. The principal difference in the different nozzle configurations is found
in the diverging supersonic-flow section, as described below. The wall surface
throughout the nozzle should be smooth and shiny to minimize friction, radiation
absorption, and convective heat transfer due to surface roughness. Gaps, holes,
sharp edges, or protrusions must be avoided.

Six different nozzle configurations are shown in Fig. 3-12 and each will be
discussed. The first three sketches show conical and bell-shaped nozzles. The
other three have a center body inside the nozzle and have excellent altitude com-
pensation. Although these last three have been ground tested, to date none of
them has flown in a production space launch vehicle. The lengths of several
nozzle types are compared in Fig. 3-13. The objectives of a good nozzle con-
figuration are to obtain the highest practical 7^, minimize inert nozzle mass, and
conserve length (shorter nozzles can reduce vehicle length, vehicle structure, and
vehicle inert mass).

Cone- and Bell-Shaped Nozzles

The conical nozzle is the oldest and perhaps the simplest configuration. It is
relatively easy to fabricate and is still used today in many small nozzles. A
theoretical correction factor k can be applied to the nozzle exit momentum of
an ideal rocket with a conical nozzle exhaust. This factor is the ratio between
the momentum of the gases in a nozzle with a finite nozzle angle 2a and the
momentum of an ideal nozzle with all gases flowing in an axial direction:

k = i(l -t-cosa) (3-34)

The variation of A with different values of a is shown in Table 3-3 for any
nozzle that has uniform mass flow per unit exit area. For ideal rockets A = 1.0.
For a rocket nozzle with a divergence cone angle of 30° (half angle a = 15°), the
exit momentum and therefore the exhaust velocity will be 98.3% of the velocity
calculated by Eq. 3- 15b. Note that the correction factor A only applies to the

Cone Contoured or Contoured or Plug or Plug or Expansion-

(15° half angle) bell-full length bell shape, aerospike full aerospike, deflection

shortened length truncated or cut off

FIGURE 3-12. Simplified diagrams of several different nozzle configurations and their flow effects.

78 NOZZLE THEORY AND THERMODYNAMIC RELATIONS

FIGURE 3-13. Length comparison of several types of nozzles. (Taken in part from Ref
3-9.)

TABLE 3-3. Nozzle Angle Correction Factor for Conical Nozzles

Nozzle Cone Divergence Half
Angle, a (deg)

Correction Factor, X

1.0000

0.9997

0.9988

0.9972

0.9951

0.9924

0.9890

0.9851

0.9830

0.9806

0.9755

0.9698

0.9636

0.9567

first term (the momentum thrust) in Eqs. 2-14, 3-29, and 3-30 and not to the
second term (pressure thrust).

A small nozzle divergence angle causes most of the momentum to be axial
and thus gives a high specific impulse, but the long nozzle has a penalty in
rocket propulsion system mass, vehicle mass, and also design complexity. A large

3.4. NOZZLE CONFIGURATIONS 79

divergence angle gives short, lightweight designs, but the performance is low.
There is an optimum conical nozzle shape and length (typically between 12 and
18° half angle) and it is usually a compromise which depends on the specific
application and flight path.

The bell-shaped or contour nozzle (see Figs. 3-12 and 3-13) is probably the
most common nozzle shape today. It has a high angle expansion section (20 to
50°) right downstream of the nozzle throat; this is followed by a gradual reversal
of nozzle contour slope so that at the nozzle exit the divergence angle is small,
usually less than a 10° half angle. It is possible to go to large divergence angles
immediately behind the throat (20 to 50°) because the high relative pressure, the
large pressure gradient, and the rapid expansion of the working fluid do not allow
separation in this region unless there are discontinuities in the nozzle contour.
The expansion in the supersonic bell nozzle is more efficient than in a simple
straight cone of similar area ratio and length, because the wall contour is designed
to minimize losses, as explained later in this section. For the past several decades
most of the nozzles have been bell shaped.

A change of flow direction of a supersonic gas in an expanding wall geometry
can only be achieved through expansion waves. An expansion wave occurs at a
thin surface within the flow, where the flow velocity increases and changes its
flow direction slightly, and where the pressure and temperature drop. These wave
surfaces are at an oblique angle to the flow. As the gas passes through the throat,
it undergoes a series of these expansion waves with essentially no loss of energy.
In the bell-shaped nozzle shown in Fig. 3-14 these expansions occur internally
in the flow between the throat and the inflection location /; the area is steadily
increasing like a flare on a trumpet. The contour angle 0, is a maximum at the
inflection location. Between the inflection point I and the nozzle exit E the flow
area is still increasing, but at a diminishing rate, allowing further gas expansion
and additional expansion waves. However, the contour of the nozzle wall is
different and the change in cross-sectional area per unit length is decreasing. The
purpose of this last segment of the contoured nozzle is to have a low divergence
loss as the gas leaves the nozzle exit plane. The angle at the exit 9^ is small,
usually less than 10°. The difference between 0, and 6e is called the turn-back
angle. When the gas flow is turned in the opposite direction (between points
I and E) oblique compression waves will occur. These compression waves are
thin surfaces where the flow undergoes a mild shock, the flow is turned, and the
velocity is actually reduced slightly. Each of these multiple compression waves
causes a small energy loss. By carefully determining the wall contour (by an
analysis that uses a mathematical tool called the method of characteristics), it is
possible to balance the oblique expansion waves with the oblique compression
waves and minimize the energy loss. The analysis leading to the nozzle contour
is presented in Chapter 20.33 of Ref. 3-3 and also in Refs. 3-8 to 3-11; it is
based on supersonic aerodynamic flow, the method of characteristics (Ref. 3-1),
and the properties of the expanding gas. Most of the rocket organizations have
computer codes for this analysis. The radius of curvature or the contour shape at

80 NOZZLE THEORY AND THERMODYNAMIC RELATIONS

Percent of length of a 15 deg half-angle
conical nozzle with same area as bell shape

FIGURE 3-14. Top sketch shows comparison of nozzle inner wall surfaces for a 15°
conical nozzle, an 80% length bell nozzle, a 60% length bell nozzle, all at an area ratio
of 25. The lengths are expressed in multiples of the throat radius r,. The middle set of
curves shows the initial angle 0, and the exit angle 0 ^ for bell nozzles as functions of the
nozzle area ratio and percent length. The bottom curves show the nozzle losses in terms
of a correction factor.

3.4. NOZZLE CONFIGURATIONS 81

the throat region have an influence on the contour of the diverging bell- shaped
nozzle section.

The length of a bell nozzle is usually given a fraction of the length of a ref-
erence conical nozzle with a 15° half angle. An 80% bell nozzle has a length
(distance between throat plane and exit plane) that is 20% shorter than a compa-
rable 15° cone of the same area ratio. Ref. 3-9 shows the original presentation
by Rao of the method of characteristics applied to shorter bell nozzles. He also
determined that a parabola was a good approximation for the bell-shaped con-
tour curve (Ref. 3-3, Section 20.33), and parabolas have actually been used in
some nozzle designs. The top part of Fig. 3-14 shows that the parabola is tan-
gent (9j) at the inflection point I and has an exit angle (0g) at point E and a
length L that has to be corrected for the curve 77. These conditions allow the
parabola to be determined by simple geometric analysis or geometric drawing.
A throat approach radius of 1.5r, and a throat expansion radius of 0.4r, were
used. If somewhat different radii had been used, the results would have been only
slightly different. The middle set of curves gives the relation between length, area
ratio, and the two angles of the bell contour. The bottom set of curves gives the
correction factors, equivalent to the A factor for conical nozzles, which are to be
applied to the thrust coefficient or the exhaust velocity, provided the nozzles are
at optimum expansions, that is, p 2 = p^.

Table 3-4 shows data for parabolas developed from this figure, which allow
the reader to apply this method and check the results. The table shows two
shortened bell nozzles and a conical nozzle, each for three area ratios. It can be
seen that as the length has been decreased, the losses are higher for the shorter
length and slightly higher for small nozzle area ratios. A 1% improvement in
the correction factor gives about 1% more specific impulse (or thrust) and this
difference can be significant in many applications. The reduced length is an

TABLE 3-4. Data on Several Bell-Shaped Nozzles

Area Ratio

Cone (15° Half Angle)

Length (100%)“

8.07

14.93

22.66

Correction factor A
80% Bell Contour

0.9829

Length“

6.45

11.94

18.12

Correction factor A
Approximate half angle at inflection
point and exit (degrees)

0.985

0.987

0.988

60% Bell Contour

Length”

4.84

9.96

13.59

Correction factor A

0.961

0.968

0.974

Approximate half angle at inflection
point and exit (degrees)

32.5/17

36/14

39/18

The length is given in dimensionless form as a multiple of the throat radius.

82 NOZZLE THEORY AND THERMODYNAMIC RELATIONS

important benefit, and it is usually reflected in an improvement of the vehicle
mass ratio. The table and Fig. 3-14 show that bell nozzles (75 to 85% length)
are just as efficient as or slightly more efficient than a longer 15° conical nozzle
(100% length) at the same area ratio. For shorter nozzles (below 70% equivalent
length) the energy losses due to internal oblique shock waves become substantial
and such short nozzles are not commonly used today.

For solid propellant rocket motor exhausts with small solid particles in the gas
(usually aluminum oxide), and for exhausts of certain gelled liquid propellants,
there is an impingement of these solid particles against the nozzle wall in the
reversing curvature section between I and E in Fig. 3-14. While the gas can be
turned by oblique waves to have less divergence, the particles (particularly the
larger particles) have a tendency to move in straight lines and hit the walls at high
velocity. The resulting abrasion and erosion of the nozzle wall can be severe,
especially with the ablative and graphite materials that are commonly used. This
abrasion by hot particles increases with turn-back angle. If the turn-back angle and
thus also the inflection angle 0, are reduced, the erosion can become acceptable.
Typical solid rocket motors flying today have values of inflection angles between
20 and 26° and turn-back angles of 10 to 15°. In comparison, current liquid rocket
engines without entrained particles have inflection angles between 27 and 50° and
turn-back angles of between 15 and 30°. Therefore the performance enhancement
caused by using a bell-shaped nozzle (high value of correction factor) is somewhat
lower in solid rocket motors with solid particles in the exhaust.

The ideal bell-shaped nozzle (minimum loss) is long, equivalent to a conical
nozzle of perhaps 10 to 12°, as seen in Fig. 3-12. It has about the same length
as a full-length aerospike nozzle. This is usually too long for reasonable vehicle
mass ratios.

Two-Step Nozzles. A two-position nozzle (with different expansion area
ratios) can give a better performance than a conventional nozzle with a fixed
single area ratio. This can be seen in Fig. 3-10; the lower area ratio nozzle
(e = 6.0) performs best at low altitudes and the higher area ratio nozzle performs
best at higher altitudes. If these two nozzles could somehow be mechanically
combined, the resulting two-position nozzle would perform closer to an ideal
nozzle which adjusts continuously to the optimum area ratio, as shown by the
thin dashed curve. When integrated over the flight time, the extra performance
can be noticeable and it has a pay-off for high-velocity missions, such as
earth-orbit injection and deep space missions. Several concepts of a bell-shaped
nozzle have evolved that achieve maximum performance at more than a single
altitude. Figure 3-15 shows three concepts for a two-step nozzle, one that
has an initial low area ratio A 2 jAt for operation at or near the earth’s surface
and a larger second area ratio that improves performance at high altitudes. See
Ref. 3-4.

The extendible nozzle requires actuators, a power supply, mechanisms for mov-
ing the extension into position during flight, and fastening and sealing devices.
It has successfully flown in several solid rocket motor nozzles and in a few liq-
uid engine applications, where it was deployed prior to ignition. Although only

3.4. NOZZLE CONFIGURATIONS 83

Extendible
nozzle with
two segments
(shown in
Fig. 8-17)

Second nozzle exit segment

chamber) after moving aft
ready for operation

Center line

Droppable insert
(mechanisms for holding,
moving, or releasing the
inserts are not shown)

Chamber

FIGURE 3-15. Simplified diagrams of three altitude-compensating two-step nozzle con-
cepts. See Ref. 3-4.

two steps are shown, there have been versions with three steps; one is shown
in Fig. 12-3. The key concerns are a reliable rugged mechanism to move the
extension into position, the hot gas seal between the nozzle sections, and the
extra weight involved. Its principal merit is the short length of the nozzle while
stored. This reduces the length of the vehicle and a considerable part of the inert
vehicle mass. A further benefit is a significant performance gain, compared to a
single nozzle of a length equal to the stored length.

The droppable insert concept avoids the moving mechanism and gas seal but
has a potential stagnation temperature problem at the joint. It requires a reliable
release mechanism, and the ejected insert creates flying debris. To date it has
only some actual test experience. See Refs. 3-4 and 3-12.

The dual-bell nozzle concept uses two shortened bell nozzles combined into
one with a bump or inflection point between them, as shown in Fig. 3-15.
During ascent it functions first at the lower area ratio, with separation occurring
at the bump. As altitude increases and the gas expands further, the flow attaches
itself downstream of this point, with the flow filling the full nozzle exit section
and operating with the higher area ratio at higher performance. There is a small

84 NOZZLE THEORY AND THERMODYNAMIC RELATIONS

performance penalty for a contour with a circular bump and concern about heat
transfer. To date there has been little flight experience with this concept.

Nozzles with Aerodynamic Boundaries

These nozzles allow for nearly optimum nozzle expansion at all altitudes. The
three nozzles shown on the right side of Fig. 3-12 offer full altitude compensa-
tion and are discussed next. Refs. 3-4 and 3-8 give more information. Altitude
compensation means that the nozzle allows nearly optimum performance at every
altitude. Complete rocket engines with either a cut-off plug nozzle or an expan-
sion/deflection nozzle have been developed and have been ground tested. Tests
with both pressurized feed systems and turbopump feed systems have been suc-
cessful. However, as of 2008, none of these rocket engines with aerodynamic
nozzle boundaries have flown in a production vehicle.

The plug nozzle or aerospike nozzle has an annular doughnut- shaped cham-
ber with an annular nozzle slot. Annular nozzle throats may cause combustion
problems. An alternate practical version has a number of individual small cham-
bers (each with low area ratio short nozzles, a round throat, and a rectangular
exit) arranged in a circle around a common plug or spike; see Fig. 8-12. The
outside aerodynamic boundary of the gas flow in the divergent section of the
nozzle is the interface between the hot gas and the ambient air; there is no outer
wall as in a conical or bell-shaped nozzle. As the external or ambient pressure is
reduced during the ascending flight, this gas boundary expands outward, causes
a change in pressure distribution on the central spike, and allows an automatic
and continuous altitude compensation. The aerospike contour with the minimum
flow losses turns out to be very long, similar in length to an optimum bell nozzle
as shown in Figs. 3-12 and 3-13. The mass flow per unit exit area is relatively
uniform over the cross section and the divergence losses are minimal.

If the central plug is cut off or truncated and the wall contour is slightly
altered, then the nozzle will be very short, as shown in Figs. 3-12, 3-13, and
3-14. It will have some internal supersonic waves and will show a small but
real loss in thrust compared to a nozzle with a full length central spike. There is
a small performance loss due to shock waves generated by the impingement of
gases from the adjacent small chambers. The pressure distribution and the heat
transfer intensity vary on the inner contoured spike wall surface and vary with
altitude. Figure 8-12 shows a typical pressure distribution over the contoured
spike surface at high and low altitudes.

The pressure in the recirculating trapped gas of the subsonic region below the
bottom plate also exerts a thrust force. The losses caused by the cut-off spike
can be largely offset by injecting a small amount of the gas flow (about 1% of
total flow) through this base plate into the recirculating region, thus enhancing
the back pressure on the base plate. The advantages of the truncated aerospike
are short length (which helps to reduce the length and mass of the flight vehicle),
full altitude compensation, no flow separation from the wall at lower altitudes,
and ease of vehicle/engine integration for certain vehicle configurations.

3.5. REAL NOZZLES 85

The linear aerospike nozzle is a variation of the round axisymmetric aerospike
nozzle. Basically, it is an unrolled version of the circular configuration. It is
explained further in Section 8.2.

In the expansion deflection nozzle (Fig. 3-12) the flow from the chamber
is directed radially outward away from the nozzle axis. The flow is turned on
a curved contour outer diverging nozzle wall. The nozzle has been shortened
and has some internal oblique shock wave losses. The hot gas flow leaving the
chamber expands around a central plug. The aerodynamic interface between the
ambient air and gas flow forms an inner boundary of the gas flow in the diverging
nozzle section. As the ambient pressure is reduced, the hot gas flow fills more
and more of the nozzle diverging section. Altitude compensation is achieved by
this change in flow boundary and by changes in the pressure distribution on the
outer walls. Actual ground tests with both liquid engines and solid motors have
not yet led to routine flight tests.

Multiple Nozzles. If a single large nozzle is replaced by a cluster of smaller
nozzles on a solid motor (with the same total thrust), then it is possible to reduce
the nozzle length. Similarly, if a single large thrust chamber of a liquid engine
is replaced by several smaller thrust chambers, the nozzle length will be shorter,
reducing the vehicle length and thus the vehicle structure and inert mass. Reaction
Motors, Inc. (RMI), in their first aircraft-rocket engine, and the Soviets pioneered
a set of four thrust chambers, each with 25% of the total thrust, assembled next to
each other and fed from the same liquid propellant feed system. This quadruple
thrust chamber arrangement has been used effectively on many large Russian
space launch vehicles and missiles. As seen in Fig. 3-13, the nozzle length of
this cluster is about 30% shorter than a single nozzle of an equivalent larger thrust
chamber. The vehicle diameter at the cluster nozzle exit is somewhat larger, the
vehicle drag is higher, and there is additional engine complexity and often more
engine mass.

3.5. REAL NOZZLES

The assumptions and simplifications listed in Section 3.1 are only approxima-
tions that allow relatively simple algorithms and simple mathematical solutions
to the analysis of real rocket nozzle phenomena. For most of these assumptions
it is possible either (1) to use an empirical correction factor (based on experi-
mental data) or (2) to develop or use a more accurate algorithm, which involves
more detailed understanding and simulation of energy losses, the physical or
chemical phenomena, and also often a more complex theoretical analysis and
mathematical treatment. Some of these approaches are mentioned briefly in this
section.

Compared to an ideal nozzle, the real nozzle has energy losses and energy that
is unavailable for conversion into kinetic energy of the exhaust gas. The principal
losses are listed below and several of these are discussed in more detail.

NOZZLE THEORY AND THERMODYNAMIC RELATIONS

1. The divergence of the flow in the nozzle exit sections causes a loss, which
varies as a function of the cosine of the divergence angle as shown by
Eq. 3-34 and Table 3-3 for conical nozzles. The losses can be reduced
for bell-shaped nozzle contours.

2. Small chamber or port area cross sections relative to the throat area or low
nozzle contraction ratios A\/A, cause pressure losses in the chamber and
reduce the thrust and exhaust velocity slightly. See Table 3-2.

3. Lower flow velocity in the boundary layer or wall friction can reduce the
effective exhaust velocity by 0.5 to 1.5%.

4. Solid particles or liquid droplets in the gas can cause losses perhaps up to
5%, as described below.

5. Unsteady combustion and oscillating flow can account for a small loss.

6. Chemical reactions in nozzle flow change gas composition and gas proper-
ties and gas temperatures, giving typically a 0.5% loss. See Chapter 5.

7. There is lower pressure and lower performance during transient operation,
for example, during start, stop, or pulsing.

8. For uncooled nozzle materials, such as fiber-reinforced plastics or carbon,
the gradual erosion of the throat region increases the throat diameter by
perhaps 1 to 6% during operation. In turn this will reduce the chamber
pressure and thrust by about 1 to 6%. This enlargement of the throat area
causes a drop in chamber compression and a slight reduction in specific
impulse usually less than 0.7%.

9. Nonuniform gas composition can reduce performance (due to incomplete
mixing, turbulence, or incomplete combustion).

10. Using real gas properties can at times change the gas composition, the
values of k and SOI, and this can cause a small loss in performance, say 0.2
to 0.7%.

11. Operation at nonoptimum nozzle expansion area ratio reduces thrust and
specific impulse. There is no loss if the vehicle only flies at the altitude
for optimum nozzle expansion (j >2 — Pi,)- If it flies with a fixed nozzle area
ratio at higher or lower altitudes, then there is a loss (during a portion of the
flight) by up to 15% in thrust compared to a nozzle with altitude compen-
sation, as can be seen in Figs. 3-7 and 3-8. It also reduces performance
by 1 to 5%.

Boundary Layer

Real nozzles have a viscous boundary layer next to the nozzle walls, where the
gas velocities are much lower than the free-stream velocities in the inviscid flow
regions. An enlarged schematic view of a boundary layer is shown in Fig. 3-16.
Immediately next to the wall the flow velocity is zero and then the boundary layer
can be considered as being built up of successive annular-shaped thin layers of
increasing velocity until the free-stream velocity is reached. The low-velocity

3.5. REAL NOZZLES 87

^ .

Boundary.

layer _
thickness"

Wall thickness

1 Vw = 0

Velocity profile

Nozzle wall

Temperature profile

FIGURE 3-16. Flow conditions at a nozzle exit lip at high altitude, showing streamlines,
boundary layer, velocity, and temperature profiles.

flow close to the wall is laminar and subsonic, but in the higher-velocity regions
of the boundary layer the flow is supersonic and can become turbulent. The
local temperature in part of the boundary layer can be substantially higher than
the free-stream temperature because of the eonversion of kinetie energy into
thermal energy as the local velocity is slowed down and as heat is created by
viscous friction. The layer right next to the wall will be cooler because of heat
transfer to the wall. The gaseous boundary layer has a profound effect on the
overall heat transfer to nozzle and chamber walls. It also has an effect on the
rocket performance, particularly in applications with relatively long nozzles with
high nozzle area ratios, where a relatively high proportion of the total mass
flow (2 to 25%) can be in the lower- velocity region of the boundary layer.
The high gradients in pressure, temperature, or density and the changes in local
velocity (direction and magnitude) influence the boundary layer. Scaling laws for
boundary layer phenomena have not been reliable.

Theoretical approaches to boundary layer performance effects can be found in
Chapters 26 to 28 of Ref. 3-1 and in Ref. 3-17. A truly satisfactory theoretical
analysis of boundary layers in rocket nozzles has not yet been developed. For-
tunately, the overall effect of boundary layers on rocket performance has been
small. For most rocket nozzles the loss seldom exceeds 1% of specific impulse.

88 NOZZLE THEORY AND THERMODYNAMIC RELATIONS

Multiphase Flow

In some propulsion systems the gaseous working fluid contains small liquid
droplets and/or solid particles that must be accelerated by the gas. They give up
heat to the gas during the expansion in a nozzle. This, for example, occurs with
solid propellants (see Chapter 13) or some gelled liquid propellants (Section 7.5),
which contain aluminum powder that forms small oxide particles in the exhaust.
It can also occur with iron oxide catalysts, or propellants containing beryllium,
boron, or zirconium.

In general, if the particles are very small (typically with diameters of 0.005
mm or less), they will have almost the same velocity as the gas and will be in
thermal equilibrium with the nozzle gas flow. Thus, as the gases give up kinetic
energy to accelerate the particles, they gain thermal energy from the particles.
As the particle diameters become larger, the mass (and thus the inertia) of the
particle increases as the cube of its diameter; however, the drag force increases
only as the square of the diameter. Larger particles therefore do not move as
fast as the gas and do not give heat to the gas as readily as do smaller particles.
The larger particles have a lower momentum than an equivalent mass of smaller
particles and they reach the nozzle exit at a higher temperature than the smaller
particles, thus giving up less thermal energy.

It is possible to derive a simple theoretical approach for correcting the perfor-
mance (/j,c, or c*) as shown below and as given in Refs. 3-13 and 3-14. It is
usually based on the following assumptions: specific heats of the gases and the
particles are constant throughout the nozzle flow, the particles are small enough
to move at essentially the same velocity as the gas and are in thermal equilibrium
with the gas, and that particles do not exchange mass with the gas (no vapor-
ization or condensation). Expansion and acceleration occur only in the gas, and
the volume occupied by the particles is negligibly small compared to the gas
volume. If the amount of particles is small, the energy needed to accelerate the
particles can also be neglected. There are no chemical reactions.

The enthalpy h, the specific volume V, the specifie heat notes k, and the gas
constant R can be expressed as functions of the particle fraction p, which is
the mass of particles (liquid and/or solid) divided by the total mass. Using the
subscripts g and 5 to refer to the gas or solid state, the following relationships
then apply:

= {1 - P)(Cp)gT + pc,T

(3-35)

(3-36)

= RJ/V,

(3-37)

= (1 - P)Rs

(3-38)

(1 - P)Cp -b PCs

(1 — P)c„ + PCg

3.5. REAL NOZZLES 89

These relations are then used in the formulas for simple one-dimensional nozzle
flow, such as Eq. 2-16, 3-15, or 3-32. The values of specific impulse or char-
acteristic velocity will decrease as p, the percent of particles, is increased. For
very small particles (less than 0.01 mm in diameter) and small values of /I (less
than 6%) the loss in specific impulse is often less than 2%. For larger particles
(over 0.015 mm diameter) and larger values of ^ this theory is not helpful and
the specific impulse can be 10 to 20% less than the 4 value without flow lag.
The actual particle sizes and distribution depend on the specific propellant, the
combustion, the particular particle material, and the specific rocket propulsion
system, and usually have to be measured (see Chapters 13 and 20). Thus adding
a metal, such as aluminum, to a solid propellant will increase the performance
only if the additional heat release can increase the combustion temperature T\
sufficiently so that it more than offsets the decrease caused by the nonexpanding
mass of the particles in the exhaust.

With very high area ratio nozzles and a low nozzle exit pressure (high altitude
or space vacuum) it is possible to condense some of the propellant ingredients
that are normally gases. As the temperature drops sharply in the nozzle, it is
possible to condense gaseous species such as H2O, CO2, or NH3 and form liquid
droplets. This causes a decrease in the gas flow per unit area and the transfer
of the latent heat of vaporization to the remaining gas. The overall effect on
performance is small if the droplet size is small and the percent of condensed
gas mass is moderate. It is also possible to form a solid phase and precipitate
fine particles of snow (H2O) or frozen fog of other species.

Other Phenomena and Losses

The combustion process is really not steady. Low- and high-frequency oscillations
in chamber pressure of up to perhaps 5% of rated value are usually considered
as smooth-burning and relatively steady flow. Gas properties {k,DJl,Cp) and flow
properties (v, V,T,p, etc.) will also oscillate with time and will not necessarily
be uniform across the flow channel. These properties are therefore only “aver-
age” values, but it is not always clear what kind of an average they are. The
energy loss due to nonuniform unsteady burning is difficult to assess theoret-
ically. For smooth-burning rocket systems they are negligibly small, but they
become significant for larger-amplitude oscillations.

The composition of the gas changes somewhat in the nozzle, chemical reac-
tions occur in the flowing gas, and the assumption of a uniform or “frozen”
equilibrium gas composition is not fully valid. A more sophisticated analysis for
determining performance with changing composition and changing gas properties
is described in Chapter 5. The thermal energy that is carried out of the nozzle
(mCpT 2 ) is unavailable for conversion to useful propulsive (kinetic) energy, as is
shown in Fig. 2-3. The only way to decrease this loss is to reduce the nozzle
exit temperature T 2 (larger nozzle area ratio), but even then it is a large loss.

When the operating durations are short (as, e.g., with antitank rockets or
pulsed attitude control rockets which start and stop repeatedly), the start and

90 NOZZLE THEORY AND THERMODYNAMIC RELATIONS

Stop transients are a significant portion of the total operating time. During the
transient periods of start and stop the average thrust, chamber pressure, or specific
impulse will be lower in value than those same parameters at steady full operating
conditions. This can be analyzed in a step-by-step process. For example, during
startup the amount of propellant reacting in the chamber has to equal the flow
of gas through the nozzle plus the amount of gas needed to fill the chamber to
a higher pressure; alternatively, an empirical curve of chamber pressure versus
time can be used as the basis of such a calculation. The transition time is very
short in small, low-thmst propulsion systems, perhaps a few milliseconds, but it
can be longer (several seconds) for large propulsion systems.

Performance Correction Factors

Additional to the theoretical cone- and bell-shaped nozzle correction factors dis-
cussed in Section 3.4., we discuss here a set of empirical correction factors. These
factors represent a variety of nonideal phenomena (such as friction, imperfect
mixing and combustion, heat transfer, chemical nonequilibrium, and two- and
three-dimensional effects) which are unavoidably present; refer to Fig. 2-3 and
Ref. 3-4. Correction factors are defined for convenience in rocket analysis and
may differ from the more conventional efficiencies — there are no “universal per-
formance correction factors.” For specific propulsion systems, where accurately
measured data are available, they allow for simple predictions of the actual perfor-
mance. For example, a velocity correction factor of 0.942 means that the velocity
or the actual specific impulse is about 94% of theoretical (the commonly accepted
value might be closer to 0.92).

Corrections factors are used by engineers to predetermine performance ahead
of testing, and for preliminary designs, informal proposals, and in health moni-
toring systems. For all these, a set of accepted or nominal values is needed to
estimate performance together with some useful formula-recipes. For the most
accurate calculations, the industry relies extensively on sophisticated computer
programs which are largely proprietary.

Correction factors are ratios of the measured or actual (subscript “a”) to
the formulated or ideal (subscript “i”) values. In the testing of propulsion sys-
tems, the combustion chamber pressure, the propellant mass flow rates, the thrust
force, and the throat and exit areas are typically measured. These measurements
yield two direct ratios, namely the thrust correction factor {t^f = Fa/Fj) and the
discharge correction factor = tha/tkj), and as well the product [{pi)a(At)a]
which has the units of force. This product enters into the formulation of two other
correction factors discussed below. The nozzle area ratio [Ca = (A 2 )a/(A,)a] in
real nozzles may be found from their measured dimensions and this ratio will
differ from the ideal calculated value as will be shown in Example 3-7.

The thrust correction factor (f^ ) is found from the ratio of thrust measurements
with the corresponding ideal values of Eq. 3-29. The discharge correction factor
{t^d) can be determined from the ratio of mass-flow-rate measurements with the
corresponding theoretical values of Eq. 3-24. Unlike incompressible flows, in

3.5. REAL NOZZLES 91

rocket propulsion systems the value of is always somewhat larger than 1.0
(up to 1.15). The actual flow rate may he larger than the ideal flow rate for the
following reasons:

1 . Incomplete combustion (a lower combustion temperature) which results in
increases of the exhaust gas densities.

2. Cooling at the walls which reduces the boundary layer temperature and
thus the average gas temperature, especially in small thrust chambers.

3. Changes in the specific heat ratio and molecular mass in an actual nozzle
which affect the flow rate and thus the discharge correction factor (see
Eq. 3-24).

The c* efficiency or fc* correction factor represents the combined effective-
ness of the combustion chamber and the injector design. It can be determined
from the ratio of measured values of [ipi)a{At)ay ,ma (from Eq. 2-18) with the
corresponding ideal value of the right-hand side of Eq. 3-32. In well-designed
combustion chambers, the value of fc* correction factor is over 95%. The
correction factor , also known as the Cp efficiency, represents the effectiveness
of the nozzle design at the operating conditions. It can be determined using mea-
sured values (from Eq. 3-31) of Fa/[ip\)a{^i)a^ with the corresponding ideal
value of Eq. 3-30. In well-designed nozzles, the value of the fcp correction
factor is above 90%.

An effective exhaust velocity correction factor [fy = (Falrna)lci\ may now be
introduced using that velocity’s definition given in Eq. 2-17 (or from Eq. 3-32,
c = c*Cf) as

Kv = KpfKd = Kc*Kcp (3-40)

This further suggests a correction factors relation, a form equivalent to Eq. 3-33,
written as fp = t,dKc*Kcp which must hold only within experimental uncertainties.

Additional useful relations may now be written. The actual specific impulse,
from Eq. 2-5, may now be calculated from

{ls)a = {ls)i{K.) (3-41)

The thermodynamic nozzle efficiency which is defined as the ratio of ideal to
the actual enthalpy changes (see Eqs. 3- 15a and 3-16) under a given pressure
ratio becomes

r]„ = (Ah)a/iAh)i = \{vlf)al\{vif)i k. (f„)2 (3-42)

The approximation sign becomes an equality when p 2 = p^. This nozzle effi-
ciency will be always less than one and represents losses inside the nozzle
(Ref. 3-17). Small nozzles being used in micropropulsion can be quite inef-
ficient because of their relatively larger frictional effects which are in proportion

92 NOZZLE THEORY AND THERMODYNAMIC RELATIONS

to the surface-to-volume ratio (some being rectangular thus having additional
sharp-corner losses).

Under &tnctly frozen-flow assumptions, the right-hand side of Eq. 3-32 leads
to a useful relation between the ideal and actual stagnation temperatures [where

(To)i ^Ti]:

(To)a/iTofl « (3-43)

With conditions at the inlet of the nozzle readily calculable, one task is to
determine the actual throat area required to pass a specified mass flow rate of
gaseous propellant and also to determine the area and the flow properties at
the nozzle exit plane. Across the nozzle heat losses modify the local stagnation
temperatures and, together with friction, the stagnation pressures (which begin as
(Pq)i = Pi and Ti in the combustion chamber). Because ideal formulations are
based on the local stagnation values at any given cross section, the challenge is
to work with the appropriate flow assumptions as shown below.

When the applicable correction factors values are available, the overall ratio
of the product of the stagnation pressure with the throat area may be inferred
using

[{At)a(P\)a]/[{At)i(pi)i] = or ^d^c* (3-44)

In order to arrive at the actual nozzle throat area from such available parameters
additional information is needed. The key question is how to relate stagnation
pressure ratios to their corresponding temperature ratios across the nozzle. When
known, a poly tropic index “n” providing a relation between gas properties is
useful (Ref. 3-17); otherwise, we may apply the isentropic relation given as
Eq. 3-7 because nozzle efficiencies are typically high (around 90%). Real noz-
zle flows are not isentropic but the net entropy change along the nozzle may be
insignificant, though noticeable decreases of the stagnation pressure and temper-
ature do exist as Example 3-7 will detail next. When dealing with real nozzles
with given expansion ratios, increases of stagnation pressure over (pi), would
be necessary to achieve a desired performance.

Example 3-7. Design a rocket nozzle to conform to the following conditions;

Chamber pressure 20.4 atm = 2.067 MPa

Atmospheric pressure 1.0 atm

Mean molecular mass of gases 21.87 kg/kg-mol

Specific heat ratio 1.23

Ideal specific impulse 230 sec (at operating conditions)

Desired thrust 1300 N

Chamber temperature 2861 K

3.5. REAL NOZZLES 93

Determine the actual exhaust velocity, the actual specific impulse, and the nozzle throat
and exit areas. Also calculate the discharge correction factor and nozzle efficiency implied
by the following correction factors which are applicable to this problem: = 0.96, fc* =

0.98, and = 0.97.

SOLUTION. The theoretical thrust coefficient can be calculated from Eq. 3-30. For
optimum conditions, p 2 = P 3 - By substituting k = 1.23 and pi/p 2 = 20.4, the thrust coef-
ficient is (Cf ); = 1.405 and (A,)/ = 4.48 cm^. These calculated values may be checked
by interpolation between the values shown in Fig. 3-6.

An actual effective exhaust velocity is not measured, but for this problem it may be
calculated from

Ca = (Ci)(iv) = (230 X 9.81)(0.98 x 0.97) = 2145 m/sec

The actual specific impulse is (Da = 2145/9.81 = 219 sec and the discharge correction
factor from Eq. 3-40

(d = Kf/Kc*Kc, = 0.96/(0.98 x 0.97) = 1.01
Because for this problem, V 2 = c, the nozzle efficiency is found from r/„ =

(f„)2 = 90%.

An approximate exit stagnation temperature may be found using Eq. 3-43,
(7b)a ^ = 2748 K. Now, an estimate of the stagnation pressure at the throat

is needed but correction factors only apply to the overall nozzle. Because the nozzle
efficiency is only 10% less than ideal, a sufficient approximation is to assign half of the
stagnation temperature drop to the region upstream of the throat, or (To)f ^ 2804 K.
Thus,

(po), « pi[2804/2861](*'"'-^) = 1.86 MPa
The throat area may now be computed from Eq. 3-44:

(At)a/(At)i = ii;F/ilCF)(Pl/(Po)t) = 1.10

As may have been expected, the actual throat area is about 10% larger than the ideal for
the same flow conditions. The actual exit area may be obtained going back to the mass
flow rate:

rh = t,pF I Ca = 0.582 kg/sec = P 2 MC/RT 2

Now, p 2 = P 3 is given and T 2 is found from

T 2 « 2748 X (0.1013/1.67)<°-22®''‘-22®) = 1631 K

where (po)a ^ 1-67 MPa is an actual stagnation pressure applicable to the exit of the
real nozzle. Finally, the exit area and area ratio are found to be. (A 2 )a 16.61 cm^ and

Ca ^ 3.37, which may be compared to the value from Fig. 3-4.

94 NOZZLE THEORY AND THERMODYNAMIC RELATIONS

Four Performance Parameters

When using values or quoting of thrust, specific impulse, propellant flow, and
other performance parameters, one must be careful to specify or qualify the
conditions under which a specific number is presented. There are at least four
sets of performance parameters and they are different in concept and value,
even when referring to the same rocket propulsion system. Each performance
parameter, such as F, 7^, c, V 2 , and/or hi, should be accompanied by a clear
definition of the conditions under which it applies. Not all the items below apply
to every one of the parameters.

(a) Chamber pressure; also, for slender chambers, the location where this pres-
sure prevails or is measured (e.g., at nozzle entrance).

(b) Ambient pressure or altitude or space (vaeuum).

(d) Nozzle shape and exit angle (see Table 3-3 and Fig. 3-12).

(e) Propellants, their composition or mixture ratio.

(f) Key assumptions and corrections made in the calculations of the theoretical
performance: for example, was frozen or shifting equilibrium used in the
analysis? (This is described in Chapter 5.)

(g) Initial ambient temperature of propellants, prior to start.

1. Theoretical performance values are defined in Chapters 2, 3, and 5 and
generally apply to ideal rockets. The conditions for an ideal nozzle were given
in Section 3.1. The theoretical performance of an ideal nozzle per se is not
used, but one that includes theoretical corrections is used. Analyses of one or
more losses can be included, and these will yield a lower theoretical perfor-
mance and also an assessment in the correction factors described in the previous
section. Most are two dimensional and correct for the chemical reactions in the
nozzle using real gas properties, and most correct for divergence. Many also
correct for one or more of the other losses mentioned above. For example, pro-
grams for solid propellant motor nozzles can include losses for throat erosion
and multiphase flow; for liquid propellant engines it may include two or more
concentric zones, each at different mixture ratios and thus with different gas
properties. Nozzle wall contour analysis with expansion and compression waves
may use a finite element analysis and/or a method of characteristics approach.
Some of the more sophisticated programs include analyses of viscous boundary
layer effects and heat transfer to the walls. Typically these computer simulation
programs are based on computer fluid dynamics finite element analyses and on
the basic Navier-Stokes relationships. Most organizations doing nozzle design
have their own computer programs, often different programs for different nozzle
designs, different thrust levels, or operating durations. Many also have simpler.

3.5. REAL NOZZLES 95

one-dimensional computer programs which may include one or more of the
above corrections; they are used frequently for preliminary estimates or informal
proposals.

2. Delivered, that is, actually measured, performance values are obtained from
static tests or flight tests of full-scale propulsion systems. Again, the conditions
should be explained (e.g., define pi,A 2 /At,p^, etc.) and the measured values
should be corrected for instrument deviations, errors, or calibration constants.
Flight test data need to be corrected for aerodynamic effects, such as drag. Often
empirical coefficients, such as the thrust correction factor, the velocity correction
factor, and the mass discharge flow correction factors are used to convert the
theoretical values of item 1 above to approximate actual values and this is often
satisfactory for preliminary estimates. Sometimes subscale thrust chambers have
been used in the development of new thrust chambers and then scale factors are
used to correct the measured data to full-scale values.

3. Performance values at standard conditions are corrected values of items
1 and 2 above. These standard conditions are generally rigidly specified by the
customer and by commonly accepted industrial practice. Usually they refer to
conditions that allow ready evaluation or comparison with reference values and
often they refer to conditions that can be easily measured and/or corrected. For
example, to allow a good comparison of specific impulse for several propellants
or rocket propulsion systems, the values are often corrected to the following
standard conditions (see Example 3-6):

(a) Pi = 1000 psia or 6.894 x 10^ Pa.

(b) p 2 = p 3 = 14.69 psia (sea level) or 1.0132 x 10^ Pa or 0.10132 Mpa.

(d) Nozzle divergence half angle a = 15° for conical nozzles, or some agreed-
upon value.

(e) Specific propellant, its design mixture ratio, and/or propellant composition.

(f) Propellant initial ambient temperature: 21°C (sometimes 20 or 25°C) or boil-
ing temperature, if cryogenic.

A rocket propulsion system is generally designed, built, tested, and delivered
in accordance with some predetermined requirements or specifications, usually in
formal documents often called the rocket engine or rocket motor specifications .
They may define the performance as shown above and they also define many
other requirements. More discussion of these specifications is given as a part of
the selection process for propulsion systems in Chapter 19.

4. Rocket manufacturers are often required by their customers to deliver
rocket propulsion systems with a guaranteed minimum performance , such as
minimum F or minimum f and/or both. The determination of this value can
be based on a nominal value (items 1, 2, or 3 above) diminished by all likely
losses, including changes in chamber pressure due to variation of pressure drops
in injector or pipelines, a loss due to nozzle surface roughness, propellant initial

96 NOZZLE THEORY AND THERMODYNAMIC RELATIONS

ambient temperatures, ambient pressure variations, and manufacturing variations
from rocket to rocket (e.g., in grain volume, nozzle dimensions, or pump impeller
diameters, etc.)- This minimum value can be determined by a probabilistic
evaluation of these losses and is then usually validated by actual full-scale static
and flights tests.

3.6. NOZZLE ALIGNMENT

When the thrust line or direction does not intersect the center of mass of a
flying vehicle, a turning moment will tend to rotate a vehicle in flight. Turning
moments are desirable and necessary for the controlled turning or attitude control
of a vehicle as is routinely done by means of the deflection of the thrust vector,
aerodynamic fins, or by separate attitude control rocket engines. However, this
turning is undesirable when its magnitude or direction is not known; this happens
when a fixed nozzle of a major propulsion system has its thrust axis misaligned.
A large high-thrust booster rocket system, even if misaligned by a very small
angle (less than 0.50°), can cause major upsetting turning moments for the firing
duration. If not corrected or compensated, such a small misalignment can cause
the flight vehicle to tumble and/or deviate from the intended flight path. For
this moment not to exceed the vehicle’s compensating attitude control capability,
it is necessary to align the nozzle axis of all propulsion systems with fixed
(nongimbal) nozzles very accurately. Normally, the geometric axis of the nozzle
diverging exit surface geometry is taken to be the thrust axis. Special alignment
fixtures are usually needed to orient the nozzle axis to be within less than ±0.25°
of the intended line to the vehicle’s center of gravity and to position the center
of a large nozzle throat to be on the vehicle centerline, say within 1 or 2 mm.
See Ref. 3-16.

There are other types of misalignments: (1) irregularities in the nozzle geom-
etry (out of round, protuberances, or unsymmetrical roughness in the surface);
(2) transient flow misalignments during start or stop; (3) uneven deflection of the
propulsion system or vehicle structure under load; and (4) irregularities in the
gas flow (faulty injector, uneven burning rate in solid propellants). For simple
unguided rocket vehicles it has been customary to rotate or spin the vehicle to
prevent the misalignment from being in one direction only or to even out the
misalignment during powered flight.

In the cramped volume of spacecraft or upper stage launch vehicles, it is
sometimes not possible to accommodate the full length of a large-area-ratio nozzle
within the available vehicle envelope. In this case the nozzles are cut off at an
angle at the vehicle surface, which allows a more compact installation. Figure
3-17 shows a diagram of two (out of four) roll control thrusters whose nozzle
exit conforms to the vehicle contour. The thrust direction of a scarfed nozzle is
no longer on the nozzle axis centerline, as it is with fully symmetrical nozzles,
and the nozzle exit flow will not be axisymmetric. Reference 3-16 shows how
to estimate the performance and thrust direction of scarfed nozzles.

PROBLEMS 97

Geometric centerline of nozzle

FIGURE 3-17. Simplified partial section of a flight vehicle showing two attitude control
thrusters with scarfed nozzles to fit a cylindrical vehicle envelope.

PROBLEMS

1. Certain experimental results indicate that the propellant gases of a liquid oxygen-
gasoline reaction have a mean molecular mass of 23.2 kg/kg-mol and a specific heat
ratio of 1.22. Compute the specific heat at constant pressure and at constant volume,
assuming a perfect gas.

2. The actual conditions for an optimum expansion nozzle operating at sea level are given
below. Calculate V 2 , Tz, and Cf- The mass flow m = 3.7 kg/sec; pi = 2.1 MPa; T\ =
2585 K; 9J1 = 18.0 kg/kg-mol; and k = 1.30.

3. A certain nozzle expands a gas under isentropic conditions. Its chamber or nozzle
entry velocity equals 90 m/sec, its final velocity 1500 m/sec. What is the change in
enthalpy of the gas? What percentage of error is introduced if the initial velocity is
neglected?

4. Nitrogen at 500°C (k = 1.38, molecular mass is 28.00) flows at a Mach number of
2.73. What are its actual and its acoustic velocity?

98 NOZZLE THEORY AND THERMODYNAMIC RELATIONS

5. The following data are given for an optimum rocket:

Average molecular mass

24 kg/kg-mol
2.533 Mpa
0.090 Mpa
2900 K

Chamber pressure
External pressure

Chamber temperature
Throat area
Specific heat ratio

0.00050 m-
1.30

Determine (a) throat velocity; (b) specific volume at throat; (c) propellant flow and
specific impulse; (d) thrust; (e) Mach number at throat.

6. Determine the ideal thrust coefficient for Problem 5 by two methods.

7. A certain ideal rocket with a nozzle area ratio of 2.3 and a throat area of 5 in.^ delivers
gases at = 1.30 and R = 66 ft-lbf/lbm-°R at a chamber pressure of 300 psia and
a constant chamber temperature of 5300°R against a back atmospheric pressure of
10 psia. By means of an appropriate valve arrangement, it is possible to throttle the
propellant flow fo the thrust chamber. Calculate and plot against pressure the following
quantities for 300, 200, and 100 psia chamber pressure: (a) pressure ratio between
chamber and atmosphere; (b) effective exhaust velocity for area ratio involved; (c)
ideal exhaust velocity for optimum and actual area ratio; (d) propellant flow; (e)
thrust; (f) specific impulse; (g) exit pressure; (h) exit temperature.

8. For an ideal rocket with a characteristic velocity c* = 1500 m/sec, a nozzle throat
diameter of 20 cm, a thrust coefficient of 1.38, and a mass flow rale of 40 kg/sec,
compute the chamber pressure, the thrust, and the specific impulse.

9. For Ihe rockel unit given in Example 3-2 compute the exhaust velocity if the nozzle
is cut off and the exit area is arbitrarily decreased by 50%. Estimate the losses in
kinetic energy and thrust and express them as a percentage of the original kinetic
energy and the original thrust.

10 . What is the maximum velocity if the nozzle in Example 3-2 was designed to expand
into a vacuum? If the expansion area ratio was 2000?

11 . Construction of a variable-area conventional axisymmetric nozzle has often been
considered to make the operation of a rocket thrust chamber take place at the optimum
expansion ratio at any altitude. Because of the enormous design difficulties of such
a mechanical device, it has never been successfully realized. Assuming that such a
mechanism can eventually be constructed, what would have to be the variation of the
area ratio with altitude (plot up to 50 km) if such a rocket had a chamber pressure
of 20 atm? Assume that k = 1 .20.

12 . Design a supersonic nozzle to operate at 10 km altitude with an area ratio of 8.0.
For the hot gas take To = 3000 K,R = 378 J/kg-K, and k = 1.3. Determine the exit
Mach number, exit velocity, and exit temperature, as well as the chamber pressure.
If this chamber pressure is doubled, what happens to the thrust and the exit velocity?
Assume no change in gas properties. How close to optimum nozzle expansion is this
nozzle?

PROBLEMS 99

13. The German World War II A-4 propulsion system had a sea-level thrust of 25,400 kg
and a chamber pressure of 1.5 Mpa. If the exit pressure is 0.084 MPa and the exit
diameter 740 mm, what is the thrust at 25,000 m?

14. Derive Eq. 3-34. {Hint: Assume that all the mass flow originates at the apex of
the cone.) Calculate the nozzle angle correction factor for a conical nozzle whose
divergence half angle is 13°.

15. For Example 3-2, determine (a) the actual thrust; (b) the actual exhaust velocity; (c)
the actual specific impulse; (d) the velocity correction factor. Assume that the thrust
correction factor is 0.985 and the discharge correction factor is 1.050.

16. An ideal rocket has the following characteristics:

Determine the critical pressure ratio, the gas velocity at the throat, the expansion area
ratio, and the theoretical nozzle exit velocity.

Answers: 0.5645; 3470 ft/sec; 14.8; and 8570 ft/sec.

17. For an ideal rocket with a characteristic velocity c* of 1220 m/sec, a mass flow rate
of 73.0 kg/sec, a thrust coefficient of 1.50, and a nozzle throat area of 0.0248 m^,
compute the effective exhaust velocity, the thrust, the chamber pressure, and the
specific impulse.

Answers: 1830 m/sec; 133,560 N; 3.590 x 10® N/m^; 186.7 sec.

18. Derive Eqs. 3-24 and 3-25.

19. An upper stage of a launch vehicle propulsion unit fails to meet expectations during
sea-level testing. This unit consists of a chamber at 4.052 MPa feeding hot propellant
to a supersonic nozzle of area ratio e = 20. The local atmospheric pressure at the
design condition is 20 kPa. The propellant has a. k = 1.2 and the throat diameter of
the nozzle is 9 cm.

(a) Calculate the ideal thrust at the design condition.

(b) Calculate the ideal thrust at the sea-level condition.

Answers: (a) 44.4 kN, (b) 34.1 kN, (c) separation in the nozzle

20. Assuming ideal flow within the propulsion unit:

(a) State all necessary conditions (realistic or not) for

Chamber pressure
Nozzle exit pressure
Specifle heat ratio
Average molecular mass
Chamber temperature

1.20

21.0 Ibm/lb-mol
4200°F

27.2 atm
3 psia

c* = c = l >2

(b) Do the above conditions result in an optimum thrust for a given pi/ps?

100 NOZZLE THEORY AND THERMODYNAMIC RELATIONS

(c) For a launch vehicle designed to operate at some intermediate earth altitude,
sketch (in absolute or relative values) how c*, c, and V 2 would vary with altitude.

21 . A rocket nozzle has been designed with A, = 19.2 in.^ and A 2 = 267 in.^ to operate
optimally at p 3 = 4 psia and produce 18,100 Ibf of ideal thrust with a chamber pres-
sure of 570 psia. It will be attached to a previously built thrust chamber that operates
at T\ = 6000°R with k = 1.25 and R = 68.75 ft-lbf/lbm°R, with a c* efficiency of
95%. But test measurements on this thrust system, at the stated pressure conditions,
yield a thrust of only 16,300 Ibf when the measured flow rate is 2.02 Ibm/sec. Find the
applicable correction factors and the actual specific impulse assuming

frozen flow throughout.

Answers: = 0.90; = 1.02; = 0.929; {Is)a = 250 sec.

SYMBOLS

lit

rii

V
vv

area, m^(ft^)

effective exhaust velocity, m/sec (ft/sec)

specific heat at constant pressure, J/kg-K (Btu/lbm-°R)

specific heat of solid, J/kg-K (Btu/lbm-°R)

specific heat at constant volume, J/kg-K (Btu/lbm-°R)

characteristic velocity, m/sec (ft/sec)

thrust coefficient

discharge coefficient (1/c*), sec/m (sec/ft)
total derivative
diameter, m (ft)
thrust, N (Ibf)

standard sea-level gravitational acceleration,

9.8066 m/sec^(32.174 ft/sec^)

enthalpy per unit mass, J/kg (Btu/lbm) (or altitude, km)

specific impulse, sec or N-sec^/kg-m (Ibf-sec/lbm)

mechanical equivalent of heat; J =4.186 J/cal in SI units or

1 Btu = 777.9 ft-lbf

specific heat ratio

length of nozzle, m (ft)

mass flow rate, kg/sec (Ibm/sec)

mach number

molecular mass, kg/kg-mol (or molecular weight, Ibm/lb-mol)
molar fraction of species i
pressure, N/m^(lbf/ft^ or Ibf/in.^)

gas constant per unit weight, J/kg-K (ft-lbf/lbm-°R) {R = R' /dR)

universal gas constant, 8314.3 J/kg mol-K (1544 ft-lb/lb mol-°R)

absolute temperature, K (°R)

velocity, m/sec (ft/sec)

specific volume, m^/kg(ft^/lbm)

propellant weight flow rate, N/sec (Ibf/sec)

REFERENCES 101

Greek Letters

half angle of divergent conical nozzle section

mass fraction of solid particles

nozzle expansion area ratio A2/A,

^Cf

thrust coefficient correction factor

fc*

c* correction factor

discharge correction factor

thrust correction factor

velocity correction factor

divergence angle correction factor for conical nozzle exit

tin

nozzle efficiency

Subscripts

actual

gas

ideal, or a particular species in a mixture

max

maximum

opt

optimum nozzle expansion

solid

sep

point of separation

throat

any plane within rocket nozzle

stagnation or impact condition

nozzle inlet or chamber

nozzle exit

atmospheric or ambient

REFERENCES

3-1. A. H. Shapiro, The Dynamics and Thermodynamics of Compressible Fluid Flow,
Vols. 1 and 2, Ronald Press Company, New York, 1953; and M. J. Zucrow and J.
D. Hoffman, Gas Dynamics, Vols. I and II, John Wiley & Sons, New York, 1976
(has section on nozzle analysis by method of characteristics).

3-2. M. J. Moran and H. N. Shapiro, Fundamentals of Engineering Thermodynamics ,
3rd ed., John Wiley & Sons, New York, 1996; also additional text, 1997.

3-3. H. H. Koelle (Ed), Handbook of Astronautical Engineering, McGraw-Hill Book
Company, New York, 1961.

3-4. G. Hagemann, H. Immich, T. Nguyen, and G. E. Dummov, “Rocket Engine Nozzle
Concepts,” Chapter 12 of Liquid Rocket Thrust Chambers: Aspects of Modeling,
Analysis and Design, V. Yang, M. Habiballah, J. Hulka, and M. Popp (Eds),
Progress in Astronautics and Aeronautics, Vol. 200, AIAA, 2004.

102

NOZZLE THEORY AND THERMODYNAMIC RELATIONS

3-5. P. Vuillermoz, C. Weiland, G. Hagemann, B. Aupoix, H. Grosdemange, and M.
Bigert, “Nozzle Design Optimization,” Chapter 13 of Liquid Rocket Thrust Cham-
bers: Aspects of Modeling, Analysis and Design, V. Yang, M. Habiballah, J. Hulka,
and M. Popp (Eds), Progress in Astronautics and Aeronautics, Vol. 200, AIAA,
2004.

3-6. “Liquid Rocket Engine Nozzles”, NASA SP-8120, 1976.

3-7. G. P. Sutton, “Flow through a Combustion Zone,” Section of Chapter 3, Rocket
Propulsion Elements, John Wiley & Sons, New York, 1956, 1963, and 1976.

3-8. J. A. Muss, T. V. Nguyen, E. J. Reske, and D. M. McDaniels, “Altitude Compen-
sating Nozzle Concepts for RLV,” AIAA Paper 97-3222, July 1997.

3-9. G. V. R. Rao, “Recent Developments in Rocket Nozzle Configurations,” ARS Jour-
nal, Vol. 31, No. 11, November 1961, pp. 1488-1494; and G. V. R. Rao, “Exhaust
Nozzle Contour for Optimum Thrust,” Jet Propulsion, Vol. 28, June 1958, pp.
377-382.

3-10. J. M. Farley and C. E. Campbell, “Performance of Several Method-of-
Characteristics Exhaust Nozzles,” NASA TN D-293 , October 1960.

3-11. J. D. Hoffman, “Design of Compressed Truncated Perfect Nozzles,” Journal of
Propulsion and Power, Vol. 3, No. 2, March-April 1987, pp. 150-156.

3-12. G. P. Sutton, Stepped Nozzle, U.S. Patent 5,779,151, 1998.

3-13. F. A. Williams, M. Barrere, and N. C. Huang, “Fundamental Aspects of Solid
Propellant Rockets,” AGARDograph 116, Advisory Group for Aerospace Research
and Development, NATO, October 1969, 783 pages.

3-14. M. Barrere, A. Jaumotte, B. Fraeijs de Veubeke, and J. Vandenkerckhove, Rocket
Propulsion, Elsevier Publishing Company, Amsterdam, 1960.

3-15. R. N. Knauber, “Thrust Misalignments of Fixed Nozzle Solid Rocket Motors,”
AIAA Paper 92-2873, 1992.

3-16. J. S. Lilley, “The Design and Optimization of Propulsion Systems Employ-
ing Scarfed Nozzles,” Journal of Spacecraft and Rockets, Vol. 23, No. 6,
November -December 1986, pp. 597-604; and J. S. Lilley, “Experimental Val-
idation of a Performance Model for Scarfed Nozzles,” Journal of Spacecraft and
Rockets, Vol. 24, No. 5, September- October 1987, pp. 474-480.

3-17. R. D. Zucker and O. Biblarz, Fundamentals of Gas Dynamics, 2nd ed., John Wiley
& Sons, Hoboken, NJ, 2002.

CHAPTER 4

FLIGHT PERFORMANCE

This chapter is an introduction to the flight performance of rocket-propelled vehi-
cles such as spacecraft, space launch vehicles, missiles, or projectiles. It presents
the subjects from a rocket propulsion point of view. Rocket propulsion systems
provide forces to a flight vehicle and cause it to accelerate (or sometimes to
decelerate), overcome drag forces, or change flight direction. Some propulsion
systems provide also torques to the flight vehicles for rotation or maneuvers. The
flight missions can be classified into several flight regimes: (1) flight within the
earth’s atmosphere (air-to-surface missiles, surface-to-surface short-range mis-
siles, surface-to-air missiles, air-to-air missiles, assisted takeoff units, sounding
rockets, or aircraft rocket propulsion systems). Refs. 4-1 and 4-2; (2) near space
environment (earth satellites, orbital space stations, or long-range ballistic mis-
siles), Refs. 4-3 to 4-9; (3) lunar and planetary flights (with or without landing
or earth return). Refs. 4-5 to 4-12; and (4) deep space exploration and sun
escape. Each is discussed in this chapter. The chapter begins with the analy-
sis of simple one-dimensional space flight and then treats more complex fight
path scenarios and various flying rocket-propelled vehicles. The appendices give
conversion factors, atmosphere properties, and a summary of key equations.

4.1. GRAVITY-FREE DRAG-FREE SPACE FLIGHT

This first simple rocket flight analysis applies to an outer space environment,
far away from any star, where there is no air (thus no drag) and essentially no
significant gravitational attraction. The flight direction is the same as the thrust
direction (along the axis of the nozzle), namely, a one-dimensional, straight-line
acceleration path; the propellant mass flow m and thus the propulsive thrust F

103

104 FLIGHT PERFORMANCE

remain constant for the propellant burning duration tp. The thrust force F has
been defined in Eq. 2-17. For a constant propellant flow the flow rate m is mpjtp,
where nip is the total usable propellant mass. From Newton’s second law and for
an instantaneous vehicle mass m and a vehicle flight velocity u

F — m dujdt (4-1)

For a rocket where the propellant flow rate is constant and the start or shut-off
periods are very brief, the instantaneous mass of the vehicle m can be expressed
as a function of the initial mass of the full vehicle mo, the propellant mass nip,
the time at power cutoff tp, and the instantaneous time f.

nip

m = mo 1 = mo

t^t\

mo tp )

= mo

1 - (1 -]V«)-

(4-2)

(4-3)

Equation 4-3 expresses the vehicle mass in a form useful for trajectory cal-
culations. The vehicle mass ratio IVR (final mass/initial mass or mf /mo ) and the
propellant mass fraction ^ have been defined by Eqs. 2-7 and 2-8. They are
related by

f = 1 — IVR = 1 — m/ /mo = mp/mo (4-4)

A definition of the various masses is shown in Fig. 4-1. The initial mass at
takeoff mo equals the sum of the useful propellant mass nip plus the empty or
final vehicle mass m^; my in turn equals the sum of the inert masses of the
engine system (such as nozzles, tanks, cases, or unused, residual propellant) plus
the guidance, control, electronics, and related equipment and the payload. The
residual propellant remaining in the propulsion system after thrust termination
is not available for combustion and is usually considered to be a part of the final
engine mass after operation. This is the liquid propellant trapped in pipe pockets,
valve cavities, and pumps or wetting the tank and pipe walls. For solid propellant
rocket motors it is the remaining unbumed solid propellant, also called slivers,
and sometimes also unbumed insulation.

For constant propellant mass flow m and a finite propellant burning time t^ the
total propellant mass mp is mtp and the instantaneous vehicle mass m = mo — liit.
Equation 4-1 can be written as

du = {F /m) dt = {cm/m)dt

{cm)dt c{mp/tp)dt ct,/tp

mo-mpt/tp moil - mpt/motp) l-^t/tp

The start period and the shutdown period are very short and can be neglected.
Integration leads to the maximum vehicle velocity at propellant burnout Up that

4.1. GRAVITY-FREE DRAG-FREE SPACE FLIGHT 105

FIGURE 4-1. Definitions of various vehicle masses. For solid propellant rocket motors
the words “rocket engine, tanks, and structures” are to be replaced by “rocket motor
nozzle, thermal insulation, and case.”

can be attained in a gravity-free vacuum. When the initial flight velocity mq is
not zero (mq 7 ^ 0) it is often called the velocity increment Am:

Am = — cln(l — ^) -t- Mo = cln(mo/m/-) -F mq (4-5)

If the initial velocity mq is assumed to be zero, then the velocity at thrust
termination

Up = Am = — cln(l — f) = — cln[mo/(mo — nip)]

= — clnlV® = cln(l/]VR)

= c \n{mo/mf) (4-6)

The symbol In stands for the natural logarithm. This Up is the maximum veloc-
ity increment Am that can be obtained in a gravity-free vacuum with constant
propellant flow, starting from rest with mq = 0. The effect of variations in c, 7^,
and ^ on the flight velocity increment is shown in Fig. 4-2. An alternate way to
write Eq. 4-6 uses e, the base of the natural logarithm:

= 1/IVR = mo/mf (4-7)

The concept of the maximum attainable flight velocity increment Am in a
gravity-free vacuum is useful in understanding the influence of the basic param-
eters. It is used in comparing one propulsion system or vehicle with another.

106 FLIGHT PERFORMANCE

Specific impulse, sec

Average effective exhaust velocity

FIGURE 4-2. Maximum vehicle flight velocity in a gravitationless, drag-free space for
different mass ratios and specific impulses (plot of Eq. 4-6). Single-stage vehicles can
have values of IVR up to about 20 and multistage vehicles can exceed 200.

one flight mission with another, or one proposed upgrade with another possible
design improvement.

From Eq. 4-6 it can be seen that the vehicle’s propellant mass fraction has a
logarithmic effect on the vehicle velocity. By increasing this ratio from 0.80 to
0.90, the interplanetary maximum vehicle velocity in gravitationless vacuum is
increased by 43%. A mass fraction of 0.80 would indicate that only 20% of the
total vehicle mass is available for structure, skin, payload, propulsion hardware,

4.2. FORCES ACTING ON A VEHICLE IN THE ATMOSPHERE 107

radios, guidance system, aerodynamic lifting surfaces, and so on; the remaining
80% is useful propellant. It requires careful design to exceed 0.85; mass fraction
ratios approaching 0.95 appear to be the probable practical limit for single-stage
vehicles and currently known materials. When the mass fraction is 0.90, then
i/m. = 0.1 and IVR = 10.0. This marked influence of mass fraction or mass
ratio on the velocity at power cutoff, and therefore also the range, not only is
true of interplanetary spaceships in a vacuum but applies to almost all types of
rocket-powered vehicles. For this reason, importance is placed on saving inert
mass on every vehicle component, including the propulsion system.

Equation 4-6 can be modihed and solved for the effective propellant mass irip
required to achieve a desired velocity increment for a given initial takeoff mass
or a final burnout mass of the vehicle. The final mass consists of the payload, the
structural mass of the vehicle, the empty propulsion system mass (which includes
residual propellant), plus a small additional mass for guidance, communications,
and control devices. Here mp = mo — m/:

mp = - 1) = mo(l - (4-8)

The flight velocity increment Up is proportional to the effective exhaust veloc-
ity c and, therefore, to the specific impulse. Thus any improvement in Is (such as
better propellants, more favorable nozzle area ratio, or higher chamber pressure)
reflects itself in improved vehicle performance, provided that such an improve-
ment does not also cause an excessive increase in rocket propulsion system inert
mass, which causes a decrease in the effective propellant fraction.

4.2. FORCES ACTING ON A VEHICLE IN THE ATMOSPHERE

The external forces commonly acting on vehicles flying in the earth’s atmosphere
are thrust, aerodynamic forces, and gravitational attractions. Other forces, such as
wind or solar radiation pressure, are usually small and generally can be neglected
for many simple calculations.

The thrust is the force produced by the power plant, such as a propeller or a
rocket. It usually acts in the direction of the axis of the power plant, that is, along
the propeller shaft axis or the rocket nozzle axis. The thrust force of a rocket
with constant mass flow has been expressed by Eq. 2-6 as a function of the
effective exhaust velocity c and the propellant flow rate rh . In many rockets the
mass rate of propellant consumption m is essentially constant, and the starting
and stopping transients are usually very short and can be neglected. Therefore,
the thrust is defined by Eq. 2-13 and is restated here with different parameters:

F = cm = cmp/tp (4-9)

As explained in Chapter 3, for a given propellant the value of the effective
exhaust velocity c or specific impulse /j depends in part on the nozzle area

108 FLIGHT PERFORMANCE

ratio and the altitude. The value of c can increase by a relatively small factor
of between 1.2 and 1.6 as altitude is increased with a maximim value in space
(vacuum).

The drag D is the aerodynamic force in a direction opposite to the flight path
due to the resistance of the body to motion in a fluid. The lift L is the aerody-
namic force acting in a direction normal to the flight path. They are expressed
as functions of the flight speed m, the mass density of the fluid (air) in which the
vehicle moves p, and a typical surface area A-.

where Cl and Co are lift and drag coefficients, respectively. For airplanes and
winged missiles the area A is the wing area. For wingless missiles or space launch
vehicles it is the maximum cross-sectional area normal to the missile axis. The lift
and drag coefficients are primarily functions of the vehicle configuration, flight
Mach number, and angle of attack, which is the angle between the vehicle axis
(or the wing plane) and the flight direction. For low flight speeds the effect of
Mach number may be neglected, and the drag and lift coefficients are functions
of the angle of attack. A typical variation of the drag and lift coefficients for a
typical supersonic missile is shown in Fig. 4-3. The values of these coefficients
reach a maximum near a Mach number of unity. For wingless vehicles the angle
of attack a is usually very small (0 < a < 1°). The density and other properties
of the atmosphere are listed in Appendix 2. The local density of the earth’s
atmosphere can vary day by day by a factor up to 2 (for altitudes of 300 to 1200
km) depending on solar activity and night-to-day temperature variations. This
introduces a major unknown in the drag. The aerodynamic forces are affected
by the flow and pressure distribution of the rocket exhaust gases, as explained in
Chapter 20.

The flight regime in the neighborhood of Mach 1 is called the transonic phase
of flight. Flere strong unsteady aerodynamic forces can develop (due to buffeting)
which are reflected in the steep rise and decrease of the coefficients seen in
Fig. 4-3. Marginal transonic load capabilities have led to structural failures of
flight vehicles.

For space launch vehicles and ballistic missiles the integrated drag loss, when
expressed in terms of Am, is typically 5 to 10% of the final ideal vehicle velocity.
This relatively low value is due to the fact that the air density is low at high
altitudes, when the velocity is high, and at low altitudes the air density is high
but the flight velocity and thus the dynamic pressure are low.

Gravitational attraction is exerted upon a flying space vehicle by all planets,
stars, the moon, and the sun. Gravity forces pull the vehicle in the direction of
the center of mass of the attracting body. Within the immediate vicinity of the
earth, the attraction of other planets and bodies is negligibly small compared to
the earth’s gravitational force. This force is the weight.

L = CijpAu^
D = CojpAu^

(4-10)

(4-11)

4.2. FORCES ACTING ON A VEHICLE IN THE ATMOSPHERE 109

0 1 2 3 4 5 6

Mach number, M

FIGURE 4-3. Variation of lift and drag coefficient with Mach number of the German
V-2 missile based on body cross-sectional area with jet off and without exhaust plume
effects at several angles of attack a.

If the variation of gravity with the geographical features and the oblate shape
of the earth are neglected, the acceleration of gravity varies inversely as the
square of the distance from the earth’s center. If Rq is the mean radius of the
spherical earth’s surface and go the acceleration on the earth’s surface at the
earth’s effective radius Rq, the gravitational attraction g changes with altitude,

g = goiRo/R?'

= go[Ro/{Ro + h)f

(4-12)

110 FLIGHT PERFORMANCE

where h is the altitude. At the equator the spherical earth’s radius is 6378.388
km and the standard value of go is 9.80665 m/sec^. At a distance as far away
as the moon, the earth’s gravity acceleration is only about 3.3 x 10“"^go- In n
more accurate analysis the value of g will vary locally with the earth’s bulge at
the equator, the high mountains, and the difference of densities of specific earth
regions.

4.3. BASIC RELATIONS OF MOTION

For a vehicle that flies within the proximity of the earth, the gravitational attrac-
tion of all other heavenly bodies may usually be neglected. Let it be assumed
that the vehicle is moving in rectilinear equilibrium flight and that all control
forces, lateral forces, and moments that tend to turn the vehicle are zero. The
trajectory is two dimensional and is contained in a fixed plane. The vehicle has
wings that are inclined to the flight path at an angle of attack a and that give a
lift in a direction normal to the flight path. The direction of flight does not nec-
essarily coincide with the direction of thrust. Figure 4-4 shows these conditions
schematically.

Let 0 be the angle of the flight path with the horizontal and \jr the angle
of the direction of thrust with the horizontal. In the direction of the flight path
the product of the mass and the acceleration has to equal the sum of all forces,
namely the propulsive, aerodynamic, and gravitational forces:

m{du/dt) = F cos(i/f — 9) — D — mg sin6> (4-13)

The acceleration perpendicular to the flight path is u{d9/dt)\ for a constant value
of u and the instantaneous radius R of the flight path it is u^/R. The equation of
motion in a direction normal to the flight velocity is

mu{d9 / dt) = F sin(i/f — 9) + L — mg cos 9 (4-14)

FIGURE 4-4. Two-dimensional free-body force diagram for flying vehicle with wings
and fins.

4.3. BASIC RELATIONS OF MOTION

111

By substituting from Eqs. 4-10 and 4-11, these two basic equations can be
solved for the accelerations as

du F Cn T

— = — cos(i/f — 0) pu^A — g sin 6 (4-15)

dt m 2m

dd F Cl 0

u — = — sin(i/f — 0) H pu^A — g cos 0 (4-16)

dt m 2m

No general solution can be given to these equations, since tp, u, Co, Cl, p,0, or
can vary independently with time, mission profile, or altitude. Also, Cl> and
Cl are functions of velocity or Mach number. In a more sophisticated analysis
other factors may be considered, such as the propellant used for nonpropulsive
purposes (e.g., attitude control or flight stability). See Refs. 4-1, 4-8, 4-11,
and 4-12 for a background of flight performance in some of the flight regimes.
Different flight performance parameters are maximized or optimized for different
rocket flight missions or flight regimes, such as An, range, orbit height and shape,
time-to-target, or altitude. Rocket propulsion systems are usually tailored to fit
specific flight missions.

Equations 4-15 and 4-16 are general and can be further simplified for various
special applications, as shown in subsequent sections. Results of such iterative
calculations of velocity, altitude, or range using the above two basic equations
often are adequate for rough design estimates. For actual trajectory analyses, nav-
igation computation, space flight path determination, or missile-firing tables, this
two-dimensional simplified theory does not permit sufficiently accurate results.
The perturbation effects, such as those listed in Section 4.4, must then be con-
sidered in addition to drag and gravity, and digital computers are necessary to
handle the complex relations. An arbitrary division of the trajectory into small
elements and a step-by-step or numerical integration to define a trajectory are
usually indicated. The more generalized three-body theory includes the gravi-
tational attraction among three masses (for example, the earth, the moon, and
the space vehicle) and is considered necessary for many space flight problems
(see Refs. 4-2 to 4-5). When the propellant flow and the thrust are not con-
stant, when the flight path is three dimensional, the form and the solution to the
equations above become more complex.

A form of Eqs. 4-15 and 4-16 can also be used to determine the actual
thrust or actual specific impulse during actual vehicle flights from accurately
observed trajectory data, such as from optical or radar tracking data. The vehicle
acceleration {du/dt) is essentially proportional to the net thrust and, by making
an assumption or measurement on the propellant flow (which usually varies in a
predetermined manner) and an analysis of aerodynamic forces, it is possible to
determine the rocket propulsion system’s actual thrust under flight conditions.

For each mission of flight one can obtain actual histories of velocities and
distances traveled and thus complete trajectories when integrating Eqs. 4-15 and
4-16. The more general case requires six equations; three for translation along
each of three perpendicular axes and three for rotation about these axes. The

112 FLIGHT PERFORMANCE

FIGURE 4-5. Simplified free-body force diagram for vehicle without wings or fins. The
force vector diagram shows the net force on the vehicle.

choice of coordinate systems and the reference points can simplify the mathe-
matical solutions (see Refs. 4-3 and 4-5).

For a wingless rocket projectile, a space launch vehicle, or a missile with
constant thmst and propellant flow, these equations can be simplified. In Fig. 4-5
the flight direction 0 is the same as the thrust direction and lift forces for a
symmetrical, wingless, stably flying vehicle can be assumed to be zero at zero
angle of attack. For a two-dimensional trajectory in a single plane (no wind
forces) and a stationary earth, the acceleration in the direction of flight is as
follows:

1 -

— g sin 0

Co^pu^A/mo
1 - Ct/tp

(4-17)

A force vector diagram in Fig. 4-5 shows the net force (by adding thrust, drag,
and gravity vectors) to be at an angle to the flight path, which will be curved.
These types of diagram form the basis for iterative trajectory numerical solutions.

The relationships in this section are for a two-dimensional flight path, one that
lies in a single plane. If maneuvers out of that plane are also made (e.g., due
to solar attraction, thmst misalignment, or wind), then the flight paths become
three dimensional and another set of equations will be needed to describe these
flights. Reference 4-1 describes equations for the motion of rocket projectiles
in the atmosphere in three dimensions. It requires energy and forces to push a
vehicle out of its flight plane. Trajectories have to be calculated accurately in
order to reach the intended flight objective and today all are done with the aid of
a computer. A good number of computer programs for analyzing flight trajecto-
ries exist and are maintained by aerospace companies or government agencies.

4.3. BASIC RELATIONS OF MOTION 113

Some are two dimensional, relatively simple, and used for making preliminary
estimates or comparisons of alternative flight paths, alternative vehicle designs,
or alternative propulsion schemes. Several use a stationary flat earth, while others
use a rotating curved earth. Three-dimensional programs also exist, are used for
more accurate flight path analyses, and include some or all significant perturba-
tions, orbit plane changes, or flying at angles of attack. As explained in Ref. 4-4,
they are more complex.

If the flight trajectory is vertical (as for a sounding rocket), then sin0 =1.0
and COS0 = 0. Equation 4-17 can be modified:

du cKitp CdIpu^A/mq

— = — Q — (4— lo)

dt 1 - );t/tp 1 -

The velocity at the end of burning can be found by integrating between the
limits of f = 0 and t = tp when u = uq and u = Up. The first two terms can
readily be integrated. The last term is of significance only if the vehicle spends
a considerable portion of its time within the atmosphere. It can be integrated
graphically or by numerical methods, and its value can be designated as BCoA/niQ
such that

1 - t,tltp

The cutoff velocity or velocity at the end of propellant burning Up is then

BCdA

Up = -c\n{\ - l;) - gtp hMo (4-19)

where mq is the initial velocity, such as may be given by a booster, g is an
average gravitational attraction evaluated with respect to time and altitude from
Eq. 4-12, and c is a time average of the effective exhaust velocity, which is a
function of altitude.

There are always a number of trade-offs in selecting the best trajectory for a
rocket projectile. Eor example, for a fixed thrust there is a trade-off between burn-
ing time, drag, payload, maximum velocity, and maximum altitude (or range).
Reference 4-2 describes the trade-offs between payload, maximum altitude, and
flight stability for a sounding rocket.

If aerodynamic forces outside the earth’s atmosphere are neglected (operate
in a vacuum) and no booster or means for attaining an initial velocity (mq = 0) is
assumed, the velocity at the end of the burning reached in a vertically ascending
trajectory will be

Up = -cln(l - O - gtp
= —c In IVR — 'gtp
= cln(l/lV«) - gtp

(4-20)

114 FLIGHT PERFORMANCE

The first term on the right side is usually the largest and is identical to Eq. 4-6.
It is directly proportional to the effective rocket exhaust velocity and is very
sensitive to changes in the mass ratio. The second term is related to the earth’s
gravity and is always negative during ascent, but its magnitude is small if the
burning time tp is short or if the flight takes place in high orbits or in space
where g is comparatively small.

For a flight that is not following a vertical path, the gravity loss is a function of
the angle between the flight direction and the local horizontal; more specifically,
the gravity loss is the integral of g sin 9 dt.

For the simplified two-dimensional case the net acceleration a for vertical
takeoff at sea level is

where a /go is the initial takeojf acceleration in multiples of the sea-level grav-
itational acceleration go, and Fo/wo is the thrust- to- weight ratio at takeoff. For
large surface-launched vehicles, this initial-thrust-to-initial-weight ratio typically
has values between 1.2 and 2.2; for small missiles (air-to-air, air-to-surface, and
surface-to-air types) this ratio is usually larger, sometimes even as high as 50 or
100. The final or terminal acceleration af of a vehicle in vertical ascent usually
occurs just before the rocket engine is shut off and/or before the propellant is
completely consumed. If drag is neglected, then

The second version of this equation applies if the powered flight path traverses
a substantial change in altitude and thus a change in the value of g . In a gravity-
free environment this equation becomes a/- /go = Fj/wf. In rockets with constant
propellant flow the final acceleration is usually also the maximum acceleration,
because the vehicle mass to be accelerated has its minimum value just before
propellant exhaustion, and for ascending rockets the thrust usually increases with
altitude. If this terminal acceleration is too large (and causes overstressing of the
structure, thus necessitating an increase in structure mass), then the thrust can be
designed to a lower value for the last portion of the burning period. In manned
flights the maximum acceleration is limited to the maximum g loading that can
be withstood by the crew.

Example 4-1. A simple single-stage rocket for a rescue flare has the following charac-
teristics. Its flight path nomenclature is shown in the accompanying sketch.

a = {Fogo/wo) - go
a/go = (fo/wo) - 1

(4-21)

(4-22)

«//^o = (Ff/wf) - 1

= iFf/mf)go - g /go

(4-23)

4.3. BASIC RELATIONS OF MOTION 115

Launch weight

Useful propellant weight

Effective specific impulse

Launch angle 0 (relative to horizontal)

Bum time tp (with constant thrust)

4.0 Ibf
0.4 Ibf
120 sec
80°

1.0 sec

The heavy line in the ascending trajectory
designates the powered portion of the
flight.

Drag may be neglected since the flight velocities are low. Assume that the acceleration
of gravity is unchanged from its sea-level value go, which then makes the propellant mass
numerically equal to the propellant weight in the EE system, or 0.4 Ibm. Also assume
that start and stop transients are short and can be ignored.

Solve for the initial and final acceleration of powered flight, the maximum trajectory
height and the time to reach maximum height, the range or horizontal distance to impact,
and the angle at propulsion cutoff.

SOLUTION. We divide the flight path into three portions: the powered flight for 1 sec,
the unpowered ascent after cutoff, and the free-fall descent. The thrust is obtained from
Eq. 2-5:

The initial accelerations along the x and y horizontal and vertical directions are, from
Eq. 4-22,

(ao)y = go[(T sin0/w) - 1] = 32.17[(48/4.0) sin 80° - 1] = 348 ft/sec^

(ao)x = go{Fcose/w) = 32.17(48/4.0) cos 80° = 67.03 ft/sec^

At thrust termination the initial flight acceleration becomes

The vertical and horizontal components of the velocity Up at the end of powered flight are
obtained from Eq. 4-20. Note that the vehicle mass has been diminished by the propellant
that has been consumed:

{Up)y = cln(mo//ny) sin0 — g^tp = 32.17 x 120 x ln(4/3.6) x 0.984 — 32.17 x 1.0
= 368 m/sec

(Up)^ = c In (mo /my) cost? = 32.17 x 120 x ln(4/3.6) x 0.1736 = 70.6 m/sec

F = hw/tp = 120 X 0.4/1.0 = 48 Ibf

116 FLIGHT PERFORMANCE

The trajectory angle with the horizontal at rocket cutoff for a dragless flight is

The final acceleration is found, using Eq. 4-22 with the final mass, as
Of = 400 mlsec^.

For the powered flight, the coordinates at propulsion burnout and Xp can be calcu-
lated from the time integration of their respective velocities. The results are

yp = ctp[\ - \n{mQ/mf)l{mQlnif - 1)] sin0 - lIlgQtj

= 32.17120[1 - ln(4/3.6)/(4/3.6 - 1)] x 0.984 - i x 32.17 x (1.0)^

= 181 ft

Xp = ctp\\ — \n{mQ / nif) / {triQ / trif — l)]cosfl

= 32.17120[1 - ln(4/3.6)/(4/3.6 - 1)] x 0.173
= 34.7 ft

The unpowered part of the trajectory reaches zero vertical velocity at its zenith. The height
gained in unpowered free flight may be obtained by equating the vertical kinetic energy
at power cutoff to its equivalent potential energy.

The maximum height or zenith location thus becomes yz = 2105 -|- 181 = 2259 ft.
What remains now is to solve the free-flight portion of vertical descent. The time for
descent from the zenith is tz = -s/^y , /go = 11-85 sec and the final vertical or impact
vertical velocity (m/)j — goE = 381 ft/sec.

The total horizontal range to the zenith is the sum of the powered and free-flight
contributions. During free flight the horizontal velocity remains unchanged at 70.6 ft/sec
because there are no accelerations (i.e., no drag, wind, or gravity component). We now
need to find the free-flight time from burnout to the zenith, which is r = (Up)y/go =
11.4 sec. The total free-flight time becomes tff = 11.4-1- 11.85 = 23.25 sec.

Now, the horizontal or total range becomes Ajc = 34.7 -|- 70.6 x 23.25 = 1676 ft.

The impact angle would be around 79°. If drag had been included, solving this problem
would have required information on the drag coefficient (Co) and a numerical solution
using Eq. 4-18. All resulting velocities and distances would turn out somewhat lower in
value. A set of flight trajectories for sounding rockets has been worked out in Ref. 4-3.

4.4. SPACE FLIGHT

Newton’s law of gravitation defines the attraction of gravitational force Fg
between two bodies in space as follows:

tan-' (368/70.6) = 79.1°

so that

go(yz - yp)

iyz -yp) = \(Up)]lg, = 1(368)2/32.17 = 2105 ft

Fg = Gm\m2/R^ = fxm2/R^

(4-24)

4.4. SPACE FLIGHT 117

Here G is the universal gravity constant (G = 6.670 x 10“"m^/kg-sec^),mi
and m 2 are the masses of the two attracting bodies (such as the earth and the
moon, the earth and a spacecraft, or the sun and a planet), and R is the distance
between their centers of mass. The earth’s gravitational constant /x is the product
of Newton’s universal constant G and the mass of the earth, mi (5.974 x 10^’^kg).
It is /X = 3.98600 x lO'^^m^/sec^.

The rocket offers a means for escaping the earth for lunar and interplanetary
travel, for escaping our solar system, and for creating a stationary or moving
station in space. The flight velocity required to escape from the earth can be
found by equating the kinetic energy of a moving body to the work necessary to
overcome gravity, neglecting the rotation of the earth and the attraction of other
celestial bodies:

imw^ = m j g dR

By substituting for g from Eq. 4-12 and by neglecting air friction the follow-
ing relation for the escape velocity Ve is obtained:

Ve = Ro

2g0

Rq 4- h

(4-25)

Here Rq is the effective earth mean radius (6374.2 km), h is the orbit altitude
above sea level, and go is the acceleration of gravity at the earth surface (9.806
m/sec). The satellite flight radius R measured from the earth’s center is R =
Ro + h. The velocity of escape at the earth’s surface is 11,179 m/sec or 36,676
ft/sec and does not vary appreciably within the earth’s atmosphere, as shown
by Fig. 4-6. Escape velocities for surface launch are given in Table 4-1 for
the sun, the planets, and the moon. Launching from the earth’s surface at escape
velocity is not practical. As a vehicle ascends through the earth’s atmosphere, it is
subject to severe aerodynamic heating and dynamic pressures. A practical launch
vehicle has to traverse the atmosphere at relatively low velocity and accelerate to
the high velocities beyond the dense atmosphere. For example, during a portion
of the Space Shuttle’s ascent, its main engines are actually throttled to a lower
thrust to avoid excessive pressure and heating. Alternatively, an escape vehicle
can be launched from an orbiting space station or from an orbiting Space Shuttle.

A rocket spaceship can become a satellite of the earth and revolve around the
earth in a fashion similar to that of the moon. Satellite orbits are usually ellip-
tical and some are circular. Low earth orbits, typically below 500 km altitude,
are designated by the letters LEO. The altitude of the orbit is usually above the
earth’s atmosphere, because this minimizes the expending of energy to overcome
the small drag which pulls the vehicle closer to the earth. The effects of the radi-
ation in the Van Allen belt on human beings and sensitive equipment sometimes
necessitate the selection of an earth orbit at low altitude.

For a circular trajectory the velocity of a satellite must be sufficiently high so
that its centrifugal force balances the earth’s gravitational attraction:

mu^ /R — mg

118 FLIGHT PERFORMANCE

15,000

10,000

5,000

Altitude, 10® ft

5 10

Altitude, 10® m

FIGURE 4-6. Orbital energy, orbital velocity, period of revolution, and earth escape
velocity of a space vehicle as a function of altitude for circular satellite orhits. It is based
on a spherical earth and neglects the earth’s oblate shape, rotation, and atmospheric drag.

For a circular orbit, the satellite velocity Us is found by using Eq. 4-12,

Mi = RoV go/ (Ro + h) = II /R (4-26)

which is smaller than the escape velocity by a factor of -Jl. The period t in
seconds of one revolution for a circular orbit relative to a stationary earth is

T = l7i{Ro + h)/us = 27t(7?o + hfl^KRo^ ) (4-27)

The energy E necessary to bring a unit of mass into a circular satellite orbit
neglecting drag consists of kinetic and potential energy, namely,

= -Rogo

Ro + 2/f
Ro + h

(4-28)

The escape velocity, satellite velocity, satellite period, and satellite orbital energy
are shown as functions of altitude in Fig. 4-6.

A satellite circulating around the earth at an altitude of 300 miles or 482.8 km
has a velocity of about 7375 m/sec or 24,200 ft/sec, circles a stationary earth in
1.63 hr, and ideally requires an energy of 3.35 x 10^ J to place 1 kg of spaceship
mass into its orbit. An equatorial satellite in a circular orbit at an altitude of

^ ^ I ?

fZ) o 3 C

m S
>

c ^

O >» D ^
.ti o fCP
cd > ciH o
h U 1^

i-H ^ V2

"o O on ~>

H ^

y o ca ^

'5 >

D g
Oh rn
00 ^

’c

-3

c3 ^

13
oC

T3 -t= ^
cd X _
Qi C

O O

^ t-i-^ ’~~!

g o ^

vo"

o”

'sO G^

00 o r-- o ^ cN

^WO'0^00^>0^0^0^^

CNl

m 00 o^ m ^ —

CNj —

-H Tj- CN uo CO 0^

'^COOOCOUOONCO^r^

-^coiriinwoco-HO-H

I/O

co*^

CO*'

o”

!Z3

c/3

;-l

1 1 I «

>»

o .2

o 0 =:

•c 2

1-H

r--

1-^

g o

goo

5 s s

g §

I— S c/5 hJ

H.2
O 3
Z S

^ ;

Qi X

r5 W

119

120 FLIGHT PERFORMANCE

6.611 earth radii (about 26,200 miles, 42,200 km, or 22,700 nautical miles) has a
period of revolution of exactly 24 hr. It will appear stationary to an observer on
earth. This is known as a synchronous satellite in geo synchronous earth orbit,
usually abbreviated as GEO. It is used extensively for communications satellite
applications. In Section 4.7 on launch vehicles we will describe how the payload
of a given space vehicle diminishes as the orbit circular altitude is increased
and as the inclination (angle between orbit plane and earth equatorial plane) is
changed. See Refs. 4-3, 4-4, 4-5, 4-6, and 4-9.

Elliptical Orbits

The circular orbit described above is a special case of the more general elliptic
orbit shown in Fig. 4-7; here the earth (or any other heavenly body around which
another body is moving) is located at one of the focal points of this ellipse. The
equations of motion may be derived from Kepler’s laws, and the elliptical orbit
can be described as follows, when expressed in polar coordinates:

(4-29)

where u is the velocity of the body in the elliptical orbit, R is the instantaneous
radius from the center of the attracting body (a vector quantity, which changes
direction as well as magnitude), a is the major axis of the ellipse, and /r is the
earth’s gravitational constant with a value of 3.986 x lO'^ m^/sec^. The symbols
are dehned in Fig. 4-7. From this equation it can be seen that the velocity Up is
a maximum when the moving body comes closest to its focal point at the orbit’s
perigee and that its velocity is a minimum at its apogee. By substituting for R

FIGURE 4-7. Elliptical orbit; the attracting body is at one of the focal points of the
ellipse.

4.4. SPACE FLIGHT 121

in Eq. 4-29, and by defining the ellipse’s shape factor e as the eccentricity of the
ellipse, e = — b^/a, the apogee and perigee velocities can be expressed as

/^(l - e)
a{\ + e)

M(1 +e)

a{\ — e)

(4-30)

(4-31)

Another property of an elliptical orbit is that the product of velocity and instan-
taneous radius remains constant for any location x or y on the ellipse, namely,
— UyRy = uR. The exact path that a satellite takes depends on the velocity
(magnitude and vector orientation) with which it is started or injected into its
orbit.

For interplanetary transfers the ideal mission can be achieved with minimum
energy in a simple transfer ellipse, as suggested originally by Hohmann (see Ref.
4-6). Assuming the planetary orbits about the sun to be circular and coplanar,
it can be demonstrated that the path of minimum energy is an ellipse tangent
to the planetary orbits as shown in Fig. 4-8. This operation requires a velocity
increment (relatively high thrast) at the initiation (planet A at fi) and another at
termination (planet B at t 2 )'- both increments are the velocity differences between
the respective circular planetary velocities and the perigee and apogee velocities
which define the transfer ellipse. The thrust levels at the beginning and end
maneuvers of the Hohmann ellipse must be high enough to give a short operating
time and the acceleration of at least 0.01 go, but preferably more. With electrical
propulsion these accelerations would be about 10“^go> the operating time would
be weeks or months, and the best transfer trajectories would be very different
from a Hohmann ellipse; they are described in Chapter 17.

Planet B at

FIGURE 4-8. Schematic diagram of interplanetary transfer paths. These same transfer
maneuvers apply when going from a low-altitude earth satellite orbit to a higher orbit.

122 FLIGHT PERFORMANCE

The departure date or the relative positions of the launch planet and the target
planet for a planetary transfer mission are critical, because the spacecraft has
to meet with the target planet when it arrives at the target orbit. The transfer
time (?2 — fi) for a Hohmann ellipse flight starting on earth is about 116 hr to
go to the moon and about 259 days to Mars. If a faster orbit (shorter transfer
time) is desired (see dashed lines in Fig. 4-8), it requires more energy than a
Hohmann transfer ellipse. This means a larger vehicle with more propellant and
a larger propulsion system or more total impulse. There also is a time window
for a launch of a spacecraft that will make a successful rendezvous. For a Mars
mission an earth-launched spacecraft may have a launch time window of more
than two months. A Hohmann transfer ellipse or a faster transfer path applies
not only to planetary flight but also to earth satellites, when an earth satellite
goes from one circular orbit to another (but within the same plane). Also, if
one spacecraft goes to a rendezvous with another spacecraft in a different orbit,
the two spacecraft have to be in the proper predetermined positions prior to the
launch to simultaneously reach their rendezvous location.

When the launch orbit (or launch planet) is not in the same plane as the target
orbit, then additional energy will be needed by applying thrust in a direction
normal to the launch orbit plane. More information is in Refs. 4-3, 4-4, 4-6,
and 4-10.

Example 4-2. A satellite is launched from a circular equatorial parking orbit at an
altitude of 160 km into a coplanar circular synchronous orbit by using a Hohmann transfer
ellipse. Assume a homogeneous spherical earth with a radius of 637 1 km. Determine the
velocity increments for entering the transfer ellipse and for achieving the synchronous
orbit at 42,200 km altitude. See Fig. 4-8 for the terminology of the orbits.

SOLUTION. The orbits are Ra = 6.531 x 10'’ m; Rb = 48.571 x 10® m. The major axis
a of the transfer ellipse is

Qte = + Rb) = 27.551 x 10® m

The orbit velocities of the two satellites in the two orbits are

UA = fp/RA = [3.986005 x 10*V6.571 x 10®]*''^ = 7788 m/sec
ub = f pIRb = 2864.7 m/sec

The velocities needed to enter and exit the transfer ellipse are

{u,,)a = Vm[(2/7?a) - (1/fl)]'''^ = 10,337 m/sec

(m/c)b = Vm[(2/7^b) - (1/a)]'''^ = 1394 m/sec

The changes in velocity going from parking orbit to ellipse and from ellipse to final
orbit are

tsuA = \(ute)A — ua\ = 2549 m/sec
Aub = \ub — (ute)B = 1471 m/sec

4.4. SPACE FLIGHT 123

Local Elliptical ballistic

FIGURE 4-9. Long-range ballistic missiles follow an elliptical free flight trajectory,
which is drag free, with the earth’s center as one of the focal points. The surface launch
is usually vertically up (not shown here) but the flight path is quickly tilted during the early
powered flight to enter into an elliptic trajectory. The hallistic range is the arc distance
on the earth’s surface. The same elliptical flight path can be used by launch vehicles for
satellites; another powered flight period occurs (called orhit injection) just as the vehicle
is at its elliptical apogee (as indicated hy the arrow), causing the vehicle to enter an orhit.

The total velocity change for the transfer maneuvers is

Awtotai = Ama + Amb = 4020 m/sec

Figure 4-9 shows the elliptical transfer trajectory of a ballistic missile or a satellite
ascent vehicle. During the initial powered flight the trajectory angle is adjusted hy the guid-
ance system to an angle that will allow the vehicle to reach the apogee of its elliptical path
exactly at the desired orhit altitude. For the ideal satellite the simplified theory assumes
that orbit injection is an essentially instantaneous application of the total impulse as the
ballistic trajectory reaches its apogee or zenith. In reality the rocket propulsion system
operates over a finite time, during which gravity losses and changes in altitude occur.

Deep Space

Lunar and interplanetary missions include circumnavigation, landing, and return
flights to the moon, Venus, Mars, and other planets. The energy necessary to

124 FLIGHT PERFORMANCE

Operation of retro rocket
to slow vehicle down
to satellite velocity (1-5%)

C 03 S* let

trajec

. . at touchdo

Vertical launch and

Retro-rock

Acceleration

maneuver to attain

Retro-rocket maneuver
to de-orbit into a landing
approach (1-5%)

interplanetary
orbit (1-10%)

Coast in orbit

FIGURE 4-10. Schematic diagram of typical powered flight maneuvers during a hypo-
thetical two-dimensional interplanetary mission with a landing. The numbers indicate
typical thrust magnitudes of the maneuvers in percent of launch takeoff thrust. This is not
drawn to scale. Heavy lines show powered flight path segments.

escape from the earth can be calculated as jtnv^ from Eq. 4-25. It is 6.26 x
10^ J/kg, which is more than that required for a satellite. The gravitational attrac-
tion of various heavenly bodies and their respective escape velocities depends
on their masses and diameters; approximate values are listed in Table 4-1. An
idealized diagram of an interplanetary landing mission is shown in Fig. 4-10.

The escape from the solar system requires approximately 5.03 x 10^ J/kg.
This is eight times as much energy as is required for escape from the earth.
There is technology to send small, unmanned probes away from the sun to outer
space; as yet there needs to be an invention and demonstrated proof of a long-
duration, novel, rocket propulsion system before a mission to the nearest star can
be achieved. The trajectory for a spacecraft to escape from the sun is either a
parabola (minimum energy) or a hyperbola. See Refs. 4-6 and 4-10.

Perturbations

This section gives a brief discussion of the disturbing torques and forces which
cause perturbations or deviations from any intended space flight path or any
satellite’s flight orbit. For a more detailed treatment of flight paths and their
perturbations, see Refs. 4-3, 4-4, and 4-13. A system is needed to measure
the satellite’s position and deviation from the intended flight path, to determine
the needed periodic correction maneuver, and then to counteract, control, and
correct them. This is called orbit maintenance ; it corrects the perturbed or altered
orbit by periodically applying small rocket propulsion forces in predetermined
directions. Typically, the corrections are performed by a set of small reaction
control thrusters which provide predetermined total impulses into the desired

4.4. SPACE FLIGHT

125

FIGURE 4-11. The regression of nodes is shown as a rotation of the plane of the orbit.
The direction of the movement will be opposite to the east-west components of the
earth’s satellite motion.

directions. These corrections are needed throughout the life of the spacecraft (for
1 to 20 years or sometimes more) to overcome the effects of the disturbances
and maintain the intended flight regime.

Perturbations can be categorized as short term and long term. The daily or
orbital period oscillating forces are called diurnal and those with long periods
are called secular.

High-altitude earth satellites (36,000 km and higher) experience perturbing
forces primarily as gravitational pull from the sun and the moon, with the forces
acting in different directions as the satellite flies around the earth. This third-body
effect can increase or decrease the velocity magnitude and change its direction.
In extreme cases the satellite can come very close to the third body, such as a
planet or one of its moons, and undergo what is called a hyperbolic maneuver
that will radically change the trajectory. This encounter can be used to increase
or decrease the energy of the satellite and intentionally change the velocity and
the shape of the orbit.

Medium- and low-altitude satellites (500 to 35,000 km) experience perturba-
tions because of the earth’s oblateness. The earth bulges in the vicinity of the
equator and a cross section through the poles is not entirely circular. Depending
on the inclination of the orbital plane to the earth equator and the altitude of the
satellite orbit, two perturbations result: (1) the regression of the nodes and (2)
shifting of the apsides line (major axis). Regression of the nodes is shown in
Fig. 4-11 as a rotation of the plane of the orbit in space, and it can be as high
as 9° per day at relatively low altitudes. Theoretically, regression does not occur
in equatorial orbits.

Figure 4-12 shows an exaggerated shift of the apsidal line, with the center of
the earth remaining as a focus point. This perturbation may be visualized as the

126

FLIGHT PERFORMANCE

FIGURE 4-12. Shifting of the apsidal line of an elliptic orbit from position 1 to 2
because of the oblateness of the earth.

movement of the prescribed elliptical orbit in a fixed plane. Obviously, both the
apogee and perigee points change in position, the rate of change being a function
of the satellite altitude and plane inclination angle. At an apogee altitude of 1000
nautical miles (n.m.) and a perigee of 100 n.m. in an equatorial orbit, the apsidal
drift is approximately 10° per day.

Satellites of modern design, with irregular shapes due to protruding antennas,
solar arrays, or other asymmetrical appendages, experience torques and forces
that tend to perturb the satellite’s position and orbit throughout its orbital life.
The principal torques and forces result from the following factors:

1 . Aerodynamic drag . This factor is significant at orbital altitudes below 500
km and is usually assumed to cease at 800 km above the earth. Reference
4-8 gives a detailed discussion of aerodynamic drag, which, in addition
to affecting the attitude of unsymmetrical vehicles, causes a change in
elliptical orbits known as apsidal drift, a decrease in the major axis, and a
decrease in eccentricity of orbits about the earth. See Refs. 4-6, -8, -12,
and -13.

2. Solar radiation. This factor dominates at high altitudes (above 800 km)
and is due to impingement of solar photons upon satellite surfaces. The
solar radiation pressure p (N/m^) on a given surface of the satellite in the
vicinity of the earth exposed to the sun can be determined as

p = 4.5x 10“®cos6»[(l -k,)cos6> + 0.67krf] (4-32)

where 9 is the angle (degrees) between the incident radiation vector and the
normal to the surface and and k^i are the specular and diffuse coefficients
of reflectivity. Typical values are 0.9 and 0.5, respectively, for ks and kd
on the body and antenna and 0.25 and 0.01, respectively, for kg and kd
with solar array surfaces. The radiation intensity varies as the square of

4.4. SPACE FLIGHT 127

the distance from the sun (see Refs. 4-4 and 4-14). The torque T on the
vehicle is given by T = pAl, where A is the projected area and I is the
offset distance between the spacecraft’s center of gravity and the center
of solar pressure. For a nonsymmetrical satellite with a large solar panel
on one side the radiation will cause a small torque, which will rotate the
vehicle.

3. Gravity gradients. Gravitational torque in spacecraft results from a vari-
ation in the gravitational force on the distributed mass of a spacecraft.
Determination of this torque requires knowledge of the gravitational field
and the distribution of spacecraft mass. This torque decreases as a function
of the orbit radius and increases with the offset distances of masses within
the spacecraft (including booms and appendages); it is most significant in
large spacecraft or space stations operating in relatively low orbits (see
Refs. 4-4 and 4-15.

4. Magnetic field . The earth’s magnetic field and any magnetic moment within
the satellite interact to produce torque. The earth’s magnetic field precesses
about the earth’s axis but is very weak (0.63 and 0.31 gauss at poles and
equator, respectively). This field is continually fluctuating in direction and
intensity because of magnetic storms and other influences. Since the field
strength decreases with \/R^ with the orbital altitude, magnetic field forces
are often neglected in the preliminary analysis of satellite flight paths (see
Ref. 4-16).

5. Internal accelerations . Deployment of solar array panels, the shifting of
liquid propellant, the movement of astronauts or other masses within the
satellite, or the “unloading” of reaction wheels can produce torques and
forces.

6. For precise low earth orbits the oblateness of the earth (diameter at equator
is slighty lerger than diameter between poles), high mountains, or earth
surface areas of different densities will perturb these orbits.

We can categorize satellite propulsion needs according to function as listed in
Table 4-2, which shows the total impulse “budget” applicable to a typical high-
altitude, elliptic orbit satellite. The control system designer often distinguishes
two different kinds of station-keeping orbit corrections needed to keep the satellite
in a synchronous position. The east-west correction refers to a correction that
moves the point at which a satellite orbit intersects the earth’s equatorial plane
in an east or west direction; it usually corrects forces caused largely by the
oblateness of the earth. The north- south correction counteracts forces usually
connected with the third-body effects of the sun and the moon.

In many satellite missions the gradual changes in orbit caused by perturbation
forces are not of concern. However, in certain missions it is necessary to com-
pensate for these perturbing forces and maintain the satellite in a specific orbit
and in a particular position in that orbit. For example, a synchronous communi-
cations satellite in a Geo-Synchronous Earth Orbit or GEO needs to maintain its
position and its orbit, so it will be able to (1) keep covering a specific area of the

128 FLIGHT PERFORMANCE

TABLE 4-2. Propulsion Functions and Total Impulse Needs of a
2000-lbm Geosynchronous Satellite with a 7-Year Life

Function

Total Impulse
(N-sec)

Acquisition of orbit

20,000

Attitude control (rotation)

4,000

Station keeping, E-W

13,000

Station keeping, N-S

270,000

Repositioning (Am, 200 ft/sec)

53,000

Control apsidal drift (third-body attraction)

445,000

Deorbit

12,700

Total

817,700

earth or communicate with the same stations on earth within its line of sight and
(2) not become a hazard to other satellites in this densely occupied synchronous
equatorial orhit. Another example is a Low Earth Orbit or LEO communications
satellite system with several coordinated satellites; here at least one satellite has
to be in a position to receive and transmit radio-frequency (RE) signals to spe-
cific points on the earth. Their orbits, and the relative positions of these several
satellites with respect to each other, need to be controlled and maintained (see
Ref. 4-3).

Orbit maintenance means applying small correcting forces and torques peri-
odically to compensate for perturbation effects; for GEO it is typically every few
months. Typical velocity increments for the orbit maintenance of synchronous
satellites require a Am between 10 and 50 m/sec per year. For a satellite mass of
about 2000 kg a 50-m/sec correction for a 10-year orbit life would need a total
impulse of about 100,000 N-sec, which corresponds to a propellant mass of 400
to 500 kg (about a quarter of the satellite mass) if done by a small monopro-
pellant or bipropellant thrust. It would require much less propellant if electrical
propulsion were used, but in some spacecraft the inert mass of the power supply
would increase. See Refs. 4-6, 4-13, and 4-14.

Mission Velocity

A convenient way to describe the magnitude of the energy requirement of a
space mission is to use the concept of the mission velocity. It is the sum of
all the flight velocity increments needed (in all the vehicle’s stages) to attain
the mission objective even though these increments are provided by different
propulsion systems. In the simplified sketch of a planetary landing mission of
Fig. 4-10, it is the sum of all the Am velocity increments shown by the heavy
lines (rocket-powered flight segments) of the trajectories. Even through some
of the velocity increments were achieved by retro-action (a negative propulsion
force to decelerate the flight velocity), these maneuvers required energy and their
absolute magnitude is counted in the mission velocity. The initial velocity from

4.4. SPACE FLIGHT 129

TABLE 4-3. Typical Estimated Space Shuttle Incremental Flight Velocity Breakdown

Ideal satellite velocity

7790 m/sec

Am to overcome gravity losses

1220 m/sec

Am to turn the flight path from the vertical

360 m/sec

Am to counteract aerodynamic drag

118 m/sec

Orbit injection

145 m/sec

Deorbit maneuver to re-enter atmosphere and aerodynamic braking

60 m/sec

Correction maneuvers and velocity adjustments

62 m/sec

Initial velocity provided by the earth’s rotation at 28.5° latitude

—408 m/sec

Total required mission velocity

9347 m/sec

the earth’s rotation (464 m/sec at the equator and 408 m/sec at a launch station at
28.5° latitude) does not have to be provided by the vehicle’s propulsion systems.
For example, the required mission velocity for launching at Cape Kennedy, bring-
ing the space vehicle into an orbit at 110 km, staying in orbit for a while, and
then entering a deorbit maneuver has the Am components shown in Table 4-3.

The required mission velocity is the sum of the absolute values of all trans-
lation velocity increments that have forces going through the center of gravity
of the vehicle (including turning maneuvers) during the flight of the mission. It
is the theoretical hypothetical velocity that can be attained by the vehicle in a
gravity-free vacuum, if all the propulsive energy of the momentum- adding thrust
chambers in all stages were to be applied in the same direction. It is useful
for comparing one flight vehicle design with another and as an indicator of the
mission energy.

The required mission velocity has to be equal to the “supplied” mission veloc-
ity, that is, the sum of all the velocity increments provided by the propulsion
systems of each of the various vehicle stages. The total velocity increment to be
“supplied” by the shuttle’s propulsion systems for the shuttle mission described
below (solid rocket motor strap-on boosters, main engines and, for orbit injec-
tion, also the increment from the orbital maneuvering system — all shown in
Fig. 1-14) has to equal or exceed 9347 m/sec. If reaction control system propel-
lant and an uncertainty factor are added, it will need to exceed 9621 m/sec. With
chemical propulsion systems and a single stage, we can achieve a space mission
velocity of 4000 to 13,000 m/sec, depending on the payload, mass ratio, vehicle
design, and propellant. With two stages it can be between perhaps 12,000 and
22,000 m/sec.

Rotational maneuvers, described later, do not change the flight velocity and
some analysts do not add them to the mission velocity requirements. Also, main-
taining a satellite in orbit against long-term perturbing forces (see prior section)
is often not counted as part of the mission velocity. However, the designers need
to provide additional propulsion capability and propellants for these purposes.
These are often separate propulsion systems, called reaction control systems.

Typical vehicle velocities required for various interplanetary missions have
been estimated as shown in Table 4-4. By starting interplanetary journeys from

130 FLIGHT PERFORMANCE

TABLE 4-4. Approximate Vehicle Mission Velocities for Typical Interplanetary
Missions

Mission

Ideal Velocity
(km/sec)

Approximate
Actual Velocity
(km/sec)

Satellite orbit around earth (no return)

7.9-10

9.1-12.5

Escape from earth (no return)

11.2

12.9

Escape from moon

2.3

2.6

Earth to moon (soft landing on moon, no

13.1

15.2

return)

Earth to Mars (soft landing)

17.5

Earth to Venus (soft landing)

Earth to moon (landing on moon and return to

15.9

17.7

earth®)

Earth to Mars (landing on Mars and return to

22.9

earth®)

“As.sumes air braking witbin atmospheres.

TABLE 4-5. Approximate Relative Payload-Mission Comparison Chart for Typical
High-Energy Chemical Multistage Rocket Vehicles

Mission

Relative Payload® (%)

Earth satellite

100

Earth escape

35-45

Earth 24-hr orbit

10-25

Moon landing (hard)

35-45

Moon landing (soft)

10-20

Moon circumnavigation (single fly-by)

30-42

Moon satellite

20-30

Moon landing and return

1-4

Moon satellte and return

8-15

Mars flyby

20-30

Mars satellite

10-18

Mars landing

0.5-3

“300 nautical miles (555.6 km) earth orbit is 100% reference.

a space satellite station, a considerable saving in this vehicle velocity can be
achieved, namely, the velocity necessary to achieve the earth-circling satellite
orbit. As the space flight objective becomes more ambitious, the mission velocity
is increased. For a given single or multistage vehicle it is possible to increase the
vehicle’s terminal velocity, but usually only at the expense of payload. Table 4-5
shows some typical ranges of payload values for a given multistage vehicle as
a percentage of a payload for a relatively simple earth orbit. Thus a vehicle
capable of putting a substantial payload into a near-earth orbit can only land a

4.5. FLIGHT MANEUVERS 131

very small fraction of this payload on the moon, since it has to have additional
upper stages, which displace payload mass. Therefore, much larger vehicles are
required for space flights with high mission velocities if compared to a vehicle
of less mission velocity but identical payload. The values listed in Tables 4-4
and 4-5 are only approximate because they depend on specific vehicle design
features, the propellants used, exact knowledge of the trajectory -time relation,
and other factors that are beyond the scope of this short treatment.

4.5. FLIGHT MANEUVERS

In this section we describe different flight maneuvers and relate them to specific
propulsion system types. See Refs. 4-5, 4-11, 4-12, 4-13, and 4-14. The three
categories of maneuvers are:

1. In translation maneuvers the rocket propulsion thrust vector goes through
the center of gravity of the vehicle. The vehicle momentum is changed
in the direction of the flight velocity. An example of several powered
(translational maneuvers) and unpowered (coasting) segments of a complex
space flight trajectory is shown in schematic, simplified form in Fig. 4-10.
To date, most maneuvers have used chemical propulsion systems.

2. In truly rotational maneuvers there is no net thrust acting on the vehicle.
These are true force couples that apply only torque. It requires four thrusters
to be able to rotate the vehicle in either direction about any one axis (two
thrusters apart, firing simultaneously, but in opposite rotational directions).
These types of maneuver are usually provided by reaction control systems.
Most have used multiple liquid propellant thrusters, but in recent years
selected space missions have used electrical propulsion.

3. A combination of categories 1 and 2, such as a misaligned thrust vector
of a large thmst chamber that does not go exactly through the center of
gravity of the vehicle. The misalignment can be corrected by changing
the vector direction of the main propulsion system (thrust vector control)
during powered flight or by applying a simultaneous compensating torque
from a separate reaction control system.

The following types of space flight maneuvers and vehicle accelerations use
rocket propulsion. All propulsion operations are usually controlled (started, mon-
itored, and stopped) by the vehicle’s guidance and control system.

a. First stage, its upper stage propulsion systems and strap-on boosters add
momentum during launch and ascent. They require rocket propulsion of
high or medium thrusts and limited durations (typically 0.7 to 8 min). To
date all have used chemical propulsion systems. They constitute the major
mass of the space vehicle and are discussed further in the next section.

132 FLIGHT PERFORMANCE

b. Orbit injection or transferring from one orbit to another requires accurately
predetermined total impulses. It can be performed by the main propulsion
system of the top stage of the launch vehicle. More often it is done by a
separate propulsion system at lower thrust levels than the upper stages in
item (a) above. Orbit injection can be a single thrust operation after ascent
from an earth launch station. If the flight path is a Hohmann transfer ellipse
(minimum energy) or a faster transfer orbit, then two thrust application
periods are necessary, one at the beginning and one at the end of the
transfer path. For injection into earth orbit, the thrust levels are typically
between 200 and 45,000 N or 50 and 1 1,000 Ibf, depending on the payload
size, transfer time, and specific orbit. If the new orbit is higher, then the
thrusts are applied in the flight direction. If the new orbit is at a lower
altitude, then the thrusts must be applied in a direction opposite to the
flight velocity vector. The transfer orbits can also be achieved with a very
low thrust level (0.001 to 1 N) using an electric propulsion system, but
the flight paths will be very different (multiloop spiral) and the transfer
duration will be much longer. This is explained in Chapter 17. Similar
maneuvers are also performed with lunar or interplanetary flight missions,
like the planetary landing mission shown schematically in Fig. 4-10.

c. Velocity vector adjustment and minor in-flight correction maneuvers are
usually performed with low-thrust, short-duration, and intermittent (puls-
ing) operations, using a reaction control system with multiple small liquid
propellant thrusters, both for translation and rotation. The vernier rockets
on a ballistic missile are used to accurately calibrate the terminal velocity
vector for improved target accuracy. The reaction control rocket systems
in a space launch vehicle will allow accurate orbit injection adjustment
maneuvers after it is placed into orbit by another, less accurate propulsion
system. Midcourse guidance-directed correction maneuvers for the trajecto-
ries of deep space vehicles fall also into this category. Propulsion systems
for orbit maintenance maneuvers , also called station-keeping maneuvers
(to overcome perturbing forces), keep a spacecraft in its intended orbit and
orbital position and are also considered to be part of this category.

d. Reentry and landing maneuvers can take several forms. If the landing
occurs on a planet that has an atmosphere, then the drag of the atmosphere
will slow down the reentering vehicle. For a multiple elliptical orbit the
drag will progressively reduce the perigee altitude and the perigee velocity
on every orbit. Landing at a precise, preplanned location requires a particu-
lar velocity vector at a predetermined altitude and distance from the landing
site. The vehicle has to be rotated into the right position and orientation,
so as to use its heat shield correctly. The precise velocity magnitude and
direction prior to entering the denser atmosphere are critical for minimiz-
ing the heat transfer (usually to the vehicle’s heat shield) and to achieve
touchdown at the intended landing site or, in the case of ballistic missiles,
the intended target. This usually requires a relatively minor maneuver (low
total impulse). If there is very little or no atmosphere (for instance, landing

4.5. FLIGHT MANEUVERS 133

on the moon or Mercury), then a reverse thrust has to be applied during
descent and touchdown. The rocket propulsion system usually has variable
thrust to assure a soft landing and to compensate for the decrease in vehi-
cle mass as propellant is consumed during descent. The U.S. lunar landing
rocket engine, for example, had a 10 to 1 thrust variation.

e. Rendezvous and docking involve both rotational and translational maneu-
vers of small reaction control thrusters. Rendezvous and its time windows
were discussed on page xxx. Docking (sometimes called lockon) is the
linking up of two spacecraft and requires a gradual gentle approach (low
thrust, pulsing node thrusters) so as not to damage the spacecraft.

f. Simple rotational maneuvers rotate the vehicle on command into a specific
angular position so as to orient or point a telescope, instrument, solar panel,
or antenna for purposes of observation, navigation, communication, or solar
power reception. Such a maneuver is also used to keep the orientation of
a satellite in a specific direction; for example, if an antenna needs to be
continuously pointed at the center of the earth, then the satellite needs to
be rotated around its own axis once every satellite revolution. Rotation
is also used to point a nozzle of the primary propulsion system into its
intended direction just prior to its start. It can also provide pulsed thrust
for achieving flight stability, or for correcting angular oscillations, that
would otherwise increase drag or cause tumbling of the vehicle. Spinning
or rolling a vehicle about its axis will improve flight stability but will
also average out the misalignment in a thrust vector. If the rotation needs
to be performed quickly, then a chemical multithruster reaction control
system is used. If the rotational changes can be done over a long period of
time, then an electrical propulsion system with multiple thrusters is often
preferred.

g. A change of plane of the flight trajectory requires the application of a thrust
force (through the vehicle center of gravity) in a direction normal to the
original plane of the flight path. This is usually performed by a propulsion
system that has been rotated (by the reaction control system) into the proper
nozzle orientation. This maneuver is done to change the plane of a satellite
orbit or when going to a planet, such as Mars, whose orbit is inclined to
the plane of the earth’s orbit.

h. Deorbiting and disposal of used or spent spacecraft is required today to
remove space debris. The spent spacecraft should not become a hazard to
other spacecraft. A relatively small thmst will cause the vehicle to go to
a low enough elliptical orbit so that atmospheric drag will cause further
slowing. In the dense regions of the atmosphere the reentering, expended
vehicle will typically break up or overheat (bum up).

i. Emergency or alternative mission. If there is a malfunction in a spacecraft
and it is decided to abort the mission, such as a premature quick return to
the earth without pursuing the originally intended mission, then some of
the rocket engines can be used for an alternate mission. For example, the

134 FLIGHT PERFORMANCE

main rocket engine in the Apollo lunar mission service module is normally
used for retroaction to attain a lunar orbit and for return from lunar orbit
to the earth; it can be used for emergency separation of the payload from
the launch vehicle and for unusual midcourse corrections during translunar
coast, enabling an emergency earth return.

Table 4-6 lists the maneuvers that have just been described, together with
some others, and shows the various types of rocket propulsion system (as men-
tioned in Chapter 1) that have been used for each of these maneuvers. The items
with a double mark “x x” have been the preferred methods in recent years. The
table omits several propulsion systems, such as solar thermal or nuclear rocket
propulsion, because these have not yet flown in a routine space mission. The
three propulsion systems on the right of Table 4-6 are electrical propulsion sys-
tems and they have very high specific impulse (see Table 2-1), which makes
them very attractive for deep space missions and for certain station-keeping jobs
(orbit maintenance). However, they can be applied only to missions with suffi-
ciently long thrust action time for reaching the desired vehicle velocity or rotation
positions with very small acceleration.

Reaction Control System

The functions of a reaction control system have been described in the previous
section on flight maneuvers. They are used for the maneuvers identified by para-
graphs c, e, and g. In some vehicle designs they are also used for tasks described
in b, part of d, and f, if the thrust levels are low.

A reaction control system (RCS), often called an auxiliary rocket propulsion
system, is needed to provide for trajectory corrections (small Am additions) as
well as correcting the rotational or attitude position of almost all spacecraft and all
major launch vehicles. If only rotational maneuvers are made, it has been called
an attitude control system. The nomenclature has not been consistent throughout
the industry or the literature.

An RCS can be incorporated into the payload stage and each of the stages of
a multiple-stage vehicle. In some missions and designs the RCS is built into only
the uppermost stage; it operates throughout the flight and provides the control
torques and forces for all the stages. For large vehicle stages the thrust level
of multiple thrusters of an RCS can be large (500 to 15,000 Ibf) and for termi-
nal stages in small satellites they can be small (0.01 to 10.0 Ibf) and they can
be pulsed as often as commanded by the vehicle flight control system. Liquid
propellant rocket engines with multiple thrusters have been used for almost all
launch vehicles and the majority of all spacecraft. Cold gas systems were used
with early spacecraft design. In the last decade an increasing number of elec-
trical propulsion systems have been used, primarily on spacecraft, as described
in Chapter 17. The life of an RCS may be short (when used on an individual
vehicle stage), or it may see use throughout the mission duration (perhaps more
than 10 years) when part of an orbiting spacecraft.

4.5. FLIGHT MANEUVERS 135

TABLE 4-6. Types of Rocket Propulsion System Commonly Used for Different Flight
Maneuvers or Application

Liquid Propellant
Rocket Engines

Solid Propellant
Rocket Motors

Electrical

Propulsion

\ Propulsion

\ System

Flight

Maneuvers and \

Applications j, \

High Thrust, Liquid Propellant
Rocket Engine, with Turbopump

Medium to Low Thrust, Liquid
Propellant Rocket Engine

Pulsing Liquid Propellant, Multiple
Small Thrusters

Large Solid Propellant Rocket
Motor, Often Segmented

Medium to Small Solid Propellant
Motors

Arc Jet, Resisto Jet

Ion Propulsion, Electromagnetic
Propulsion

Pulsed Plasma Jet

Launch vehicle booster

X X

Strap-on motor/engine

X X

Upper stages of launch

XX XX

X XX

vehicle

Satellite orbit injection and

X X

transfer orbits

Flight velocity adjustments,

X XX

X X

flight path corrections.

orbit changes

Orbit/position maintenance,

X X

XXX

rotation of spacecraft

Docking of two spacecraft

X X

Reentry and landing.

X X

emergency maneuvers

Deorbit

X X

Deep space, sun escape

X X

Tactical missiles

X X

Strategic missiles

XXX

X X

Missile defense

XX XX

Legend : x = in use : x x = preferred for use.

The vehicle attitude has to he controlled about three mutually perpendicular
axes, each with two degrees of freedom (clockwise and counterclockwise rota-
tion), giving a total of six degrees of rotational freedom. Pitch control raises or
lowers the nose of the vehicle, yaw torques induce a motion to the right or the
left side, and roll torques will rotate the vehicle about its axis, either clockwise
or counterclockwise. In order to apply a true torque it is necessary to use two
thrust chambers of exactly equal thrust and equal start and stop times, placed an
equal distance from the center of mass. Figure 4-13 shows a simple spherical

136 FLIGHT PERFORMANCE

FIGURE 4-13. Simplified attitude control system diagram for spacecraft. It requires
12 thrusters (identified as x, y, z) to allow the application of pure torques about three
perpendicular axes. The four unlabeled thrusters are needed for translation maneuvers
along the z axis. They are shown here in four clusters.

spacecraft attitude control system; thrusters x — x' or x' — x apply torques that
rotate about the X axis. There is a minimum of 12 thrusters in this system, but
some spacecraft with geometrical or other limitations on the placement of these
nozzles or with provisions for redundancy may actually have more than 12. The
same system can, by operating a different set of nozzles, also provide transla-
tion forces; for example, if one each of the thrust units x and x' were operated
simultaneously, the resulting forces would propel the vehicle in the direction of
the Y axis. With clever design it is possible to use fewer thrusters, but they will
usually not provide a pure torque.

An RCS usually contains the following major subsystems: (1) sensing devices
for determining the attitude, velocity, and position of the vehicle with respect to a
reference direction at any one time, such as provided by gyroscopes, star-trackers,
or radio beacons; (2) a control-command system that compares the actual space
and rotary position with the desired or programmed position and issues command
signals to change the vehicle position within a desired time period; and (3) devices
for changing the angular position, such as a set of high-speed gyroscopic wheels
or a set of attitude control thrust-providing devices. See Refs. 4-13 and 4-14.

4.6. EFFECT OF PROPULSION SYSTEM ON VEHICLE PERFORMANCE 137

A precise attitude angular correction can also be achieved by the use of an
inertial or high-speed rotating reaction wheel, which applies torque when its
rotational speed is increased or decreased. While these wheels are quite simple
and effective, the total angular momentum change they can supply is generally
small. By using a pair of supplementary attitude control thrust rocket units it
is possible to unload or respin each wheel so it can continue to supply small
angular position corrections as needed.

The torque T of a pair of thrust chambers of thrust F and a separation distance
I is applied to give the vehicle with an angular or rotational moment of inertia
Ma an angular acceleration of magnitude a:

T = Fl= Mc,a (4-33)

For a cylinder of equally distributed mass Ma = and for a homogeneous
sphere it is Ma = jmr . The largest possible practical value of moment arm I
will minimize the thrust and propellant requirements. If the angular acceleration
is constant over a time period t, the vehicle will move at an angular speed co and
through a displacement angle 9, namely

CO = at and 9 = ^at^ (4-34)

Commonly a control system senses a small angular disturbance and then com-
mands an appropriate correction. For this detection of an angular position change
by an accurate sensor it is actually necessary for the vehicle to undergo a slight
angular displacement. Care must be taken to avoid overcorrection and hunting
of the vehicle position or the control system. This is one of the reasons many
spacecraft require extremely short multiple pulses (0.010 to 0.030 sec) and low
thrust (0.01 to 100 N) (see Refs. 4-11, 4-13, and 4-14).

Reaction control systems can be characterized by the magnitude of the total
impulse, the number, thrust level, and direction of the thrusters, and their duty
cycles. The duty cycle refers to the number of thrust pulses, their operating times,
the times between thrust applications, and the timing of these short operations
during the mission operating period. For a particular thruster, a 30% duty cycle
means an average active cumulative thrust period of 30% during the propul-
sion system’s flight duration. These propulsion parameters can be determined
from the mission, the guidance and control approach, the desired accuracy, flight
stability, the likely thrust misalignments of the main propulsion systems, the
three-dimensional flight path variations, the perturbations to the trajectory, and
several other factors. Some of these parameters are often difficult to determine.

4.6. EFFECT OF PROPULSION SYSTEM ON VEHICLE PERFORMANCE

This section gives several methods for improving flight vehicle performance.
Most of these enhancements, listed below, are directly influenced by the flight

138 FLIGHT PERFORMANCE

mission and by the selection or design of the propulsion system. A few of the
flight vehicle performance improvements do not depend on the propulsion system.
Most of those listed below apply to all missions, but some are peculiar to some
missions only.

1 . The ejfective exhaust velocity c and the specific impulse Is usually have a
direct effect on the vehicle’s flight performance. The vehicle’s final velocity
increment Am can be inceased by a higher This can be done by using
a more energetic chemical propellant (see Chapters 7 and 12), by a higher
chamber pressure, and, for upper stages operating at high altitudes, also
by a larger nozzle area ratio. Electrical propulsion (higher fi) can enhance
vehicle performance, but, as explained later, the very low thrusts do limit
this type to certain space missions. See Chapter 17.

2. The mass ratio mo/m/ has a logarithmic effect. It can be increased in several
ways. One way is by reducing the final mass mf, which consists of the inert
hardware plus the nonusable, residual propellant mass. Reducing the inert
mass implies lighter structures, smaller payloads, lighter guidance/control
devices, or less unavailable residual propellant; this means going to stronger
structural materials at higher stresses, more efficient power supplies, or
smaller electronic packages. During design there is always great emphasis
to reduce all hardware masses and the residual propellants to their practical
minima. Another way is to increase the initial vehicle mass, and use a higher
thrust and more propellant, but with a smaller increase in the structure or
inert propulsion system masses.

3. Reducing the burning time (i.e., increasing the thrust level) will reduce the
gravitational loss in some applications. However, the higher acceleration
usually requires more structural and propulsion system mass, which in turn
causes the mass ratio to be less favorable.

4. The drag, which can be considered as a negative thrust, can be reduced
in at least four ways. The drag has several components: (a) The form drag
depends on the aerodynamic shape. A slender pointed nose or sharp, thin
leading edges of fins or wings have less drag than a stubby, blunt shape,
(b) A vehicle with a small cross-sectional area has less drag. A propulsion
design that can be packaged in a long, thin shape will be preferred, (c) The
drag is proportional to the cross-sectional or frontal vehicle area. A higher
propellant density will decrease the propellant volume and therefore will
allow a smaller cross section, (d) The skin drag is caused by the friction of
the air flowing over all the vehicle’s outer surfaces. A smooth contour and
a polished surface are usually better. The skin drag is also influenced by
the propellant density, because it gives a smaller volume and thus a lower
surface area, (e) The base drag is the fourth component; it is a function of
the local ambient air pressure acting over the surface of the vehicle’s base
or bottom plate. It is influenced by the nozzle exit design (exit pressure),
the discharge of turbine exhaust gases, and the geometry of the vehicle
base design. It is discussed further in Chapter 20.

4.7. FLIGHT VEHICLES 139

5. The length of the propulsion nozzle often is a significant part of the overall
vehicle or stage length. As was described in Chapter 3, there is an optimum
nozzle contour and length, which can be determined by trade-off analysis.
A shorter nozzle length or multiple nozzles on the same propulsion system
allow a somewhat shorter vehicle; on many designs this implies a somewhat
lighter vehicle structure and a slightly better vehicle mass ratio.

6. The final vehicle velocity at propulsion termination can be increased by
increasing the initial velocity uq. By launching a satellite in an eastward
direction the rotational speed of the earth is added to the final satellite
orbital velocity. This tangential velocity of the earth is about 464 m/sec or
1523 ft/sec at the equator and the Sea Launch from a ship on the equator
takes full advantage of this velocity increment. For an easterly launch at
John F. Kennedy Space Center (latitude of 28.5° north) this extra velocity
is about 408 m/sec or 1340 ft/sec. Conversely, a westerly satellite launch
has a negative initial velocity and thus requires a higher-velocity increment.
Another way to increase u is to launch a spacecraft from a satellite or an
aircraft, which increases the initial vehicle velocity and allows launching
in the desired direction, or to launch an air-to-surface missile from an
airplane. An example is the Pegasus three-stage space launch vehicle, which
is launched from an airplane.

7. For vehicles that fly in the atmosphere it is possible to increase the range
when aerodynamic lift is used to counteract gravity and reduce gravity
losses. Using a set of wings or flying at an angle of attack increases the
lift, but it also increases the drag. This lift can also be used to increase the
maneuverability and trajectory flexibility.

8. When the flight velocity u is close to the rocket’s effective exhaust velocity
c, the propulsive efficiency is the highest (Eq. 2-23) and more of the
rocket exhaust gas energy is transformed into the vehicle’s flight energy.
Trajectories where u is close in value to c for a major portion of the flight
therefore need less propellant.

Several of these influencing parameters can be optimized. Therefore, for every
mission or flight application there is an optimum propulsion system design and
the propulsion parameters that define the optimum condition are dependent on
vehicle or flight parameters.

4.7. FLIGHT VEHICLES

As mentioned, the vast majority of rocket-propelled vehicles are simple, single
stage, and use solid propellant rocket motors. Most are used in military applica-
tions, as described in the next section. This section discusses more sophisticated
multistage space launch vehicles and mentions others, such as large ballistic mis-
siles (often called strategic missiles) and some sounding rockets. All have some

140 FLIGHT PERFORMANCE

intelligence in their guidance and navigation system. The total number of multi-
stage rocket vehicles produced worldwide in the last few years has been between
100 and 220 per year.

A single stage to orbit (LEO) is limited in the payload it can carry. Figure 4-2
shows that a high-performance single-stage vehicle with a propellant fraction of
0.95 and an average 4 of 400 sec can achieve an ideal terminal velocity of about
12,000 m/sec without payload. If the analysis includes drag and gravity forces, a
somewhat higher value of 4, maneuvers in the trajectory, and an attitude control
system, it is likely that the payload would be between 0.2 and 1 .4% of the gross
takeoff mass, depending on the design. For a larger percentage of payload, and
for ambitious missions, we use vehicles with two or more stages as described
here.

Multistage Vehicles

Multistep or multistage rocket vehicles permit higher vehicle velocities, more
payload for space vehicles, and improved performance for long-range ballistic
missiles. After the useful propellant is fully consumed in a particular stage,
the remaining empty mass of that expended stage is dropped from the vehicle
and the operation of the propulsion system of the next step or stage is started.
The last or top stage, which is usually the smallest, carries the payload. The
empty mass of the expended stage or step is separated from the remainder of
the vehicle, because it avoids the expenditure of additional energy for further
accelerating a useless mass. As the number of steps is increased, the initial
takeoff mass can be decreased; but the gain in a smaller initial mass becomes less
apparent when the total number of steps is large. Actually, the number of steps
chosen should not be too large, because the physical mechanisms become more
numerous, complex, and heavy. The most economical number of steps is usually
between two and six, depending on the mission. Several different multistage
launch vehicle configurations have been used successfully and four are shown in
Fig. 4-14. Most are launched vertically, but a few have been launched from an
airplane, such as the three-stage Pegasus space vehicle. See Example 4-3.

The payload of a multistage rocket is essentially proportional to the takeoff
mass, even though the payload is only a very small portion of the initial mass. If
a payload of 50 kg requires a 6000-kg multistage rocket, a 500-kg payload would
require a 60,000-kg rocket unit with an identical number of stages and a similar
configuration with the same payload fraction and the same propellants. When the
operation of the upper stage is started, immediately after thrust termination of
the lower stage, then the total ideal velocity of a multistage vehicle of tandem
or series-stage arrangement is simply the sum of the individual stage velocity
increments. For n stages in series (one on top of each other) the final velocity
increment Auf is

(4-35)

4.7. FLIGHT VEHICLES 141

Sustainer stage
(contains propellant
for booster thrust)

Second

stage

First stage
or booster

/— Dropable booster
\ engine ring package
\ (without propellant)

Partial staging

Sustainer

Parallel staging Piggy-back

staging

FIGURE 4-14. Simplified schematic sketches of four geometric configurations for
assembling individual stages into a launch vehicle.

The individual velocity increments are given by Eq. 4-6. For the simplified case
of a vacuum flight in a gravity-free field this can be expressed as

Am/- = Cl ln(l /IVRi) -|- C 2 ln(l/]VR 2 ) + C 3 ln(l/]V« 3 ) -f ■ ■ ■ (4-36)

This equation defines the maximum velocity an ideal tandem multistage vehi-
cle can attain in a gravity-free vacuum environment. For more accurate actual
trajectories the individual velocity increments can be determined by integrating
Eqs. 4-15 and 4-16, which consider drag and gravity losses. Other losses or
trajectory perturbations can also be included, as mentioned earlier in this chapter.
Such an approach requires numerical solutions.

For two- or three-stage vehicles the overall vehicle mass ratio (initial mass at
takeoff to final mass of last stage) can reach values of over 100 (corresponding

142 FLIGHT PERFORMANCE

to an equivalent single-stage propellant mass fraction ^ of 0.99). Figure 4-2
can be thus divided into regions for single- and tandem multistage vehicles.
Equation 4-36 does not apply to parallel or partial staging as identified in
Fig. 4-14. For stages where more than one propulsion system is operating at
the same time and producing thrust in the same direction, the effective specific
impulse and nozzle exhaust velocity is given by Eqs. 10-12 to 10-14.

The first sketch in Fig. 4-14 depicts a very common configuration and the
stages are stacked vertically on top of each other, as in the Minuteman long-
range missile or the Russian Zenit (Zenith) launch vehicle.* Partial staging was
used on the early versions of the U.S. Atlas vehicle. It allows all engines to
be started at launching, thus avoiding an altitude start of the sustainer engine,
which was unknown in those early days. Fiquid propellant rocket engines can
be shut off on the launch stand if a failure is sensed prior to lift-off. The two
Atlas booster engines arranged in a doughnut-shaped assembly are dropped off
in flight. The third sketch has two or more separate booster “strap-on” stages
attached to the bottom stage of a vertical configuration (they can be either solid
or liquid propellants) and this allows an increase in vehicle performance. The
piggy-back configuration concept on the right is used on the Space Shuttle. The
two large solid rocket motor boosters are not shown.

Stage Separation

It usually takes a finite time for the termination of the lower stage propulsion
system to go from full thrust to essentially zero thrust (typically 1 to 3 sec for
large thrust values). In some multistage flight vehicles (with stage separation
devices) there often was a further short delay (say 4 to 10 sec) to achieve a
respectable separation distance between the upper and the lower stage, before
the firing of the upper stage propulsion system was initiated. This was done in
order to prevent blow-back of damaging hot flames onto the upper stage. Also
the upper stage engine start-up was not instantaneous but required one or more
seconds. During this cumulative delay of several seconds the earth’s gravity pull
continues to diminish the vehicle’s upward velocity, causing a reduction of the
flight velocity by perhaps 20 to 500 ft/sec (7 to 160 m/sec). A scheme called
hot staging has been used to diminish this velocity loss and shorten the staging
time interval. The upper stage propulsion system is actually started at low but
increasing thrust before the lower stage propulsion system has been fully shut off
or well before it reaches essentially zero thrust. There are special flame-resistant
ducts in the interstage structure to allow the flame or hot exhaust gases of the
upper stage engine to be safely discharged and deflected symmetrically prior to
and immediately after the actual separation of the stages without harming the
functional hardware of the vehicle. This hot staging scheme has been used on
large multistage vehicles, such as the Titan II in the United States and certain

*The three-stage Zenit space launch vehicle is assembled in the Ukraine with Russian rocket engines.
It is transported to the United States, loaded on a special floating launch platform (Sea Launch
Program), and launched by a team that is headed by the Boeing Company.

4.7. FLIGHT VEHICLES 143

launch vehicles in China and the Soviet Union, because it does improve the flight
performance.

For multistage vehicles the stage mass ratios, thrust levels, propulsion dura-
tions, and location or travel of the center of gravity of the stages are usually
optimized, often using a complex trajectory computer program. The high specific
impulse rocket engine (e.g., using hydrogen-oxygen propellants) is normally
employed in upper stages of space launch vehicles, because a small increase in
specific impulse can be more effective there than in lower stages.

Example 4-3. A two-stage exploration vehicle is launched from a high-orbit satellite
into a gravity-free vacuum trajectory. The following notation is used and explained in the
accompanying diagram as well as in Fig. 4-1:

mo = initial mass of vehicle (or stage) at launch
mp = useful propellant mass of stage
nii = initial mass of stage(s)

nif = final mass after rocket operation; it includes the empty propulsion system with its
residual propellant, the vehicle structures plus the propulsion system with control,
guidance and payload masses

mpi = payload mass; it can include guidance, control and communications equipment,
antennas, scientific instruments, military equipment, research apparatus, power
supply, solar panels, sensors, etc.

144 FLIGHT PERFORMANCE

Subscripts 1 and 2 refer to the first and second stages. The following data are given:

Flight velocity increment in gravity-free vacuum 4700 m/sec
Specific impulse (each stage) 310 sec

Initial takeoff launch vehicle mass 4500 kg

Propellant mass fraction, t; (each stage) 0.88

Determine the payload for the following cases:

1. When the two propulsion system or stage masses are equal [(m,)i = (WiOa]-

2. When the mass ratios of the two stages are equal [(my/mo)i = [nif /mo) 2 ].

SOLUTION. The following relationships apply to both cases. The takeoff mass or launch
mass can be divided into three parts, namely, the two propulsion stages and then payload

(mo) = (m;)i + (m,)2 + (?«)pl

The propellants are 88 % of their propulsion system mass, and this is representative of
a number of stages.

(mp)i = 0 . 88 (m;)i and (m ^)2 = 0 . 88 (m ,)2

The nozzle exit area ratio and the chamber pressure are the same in both stages and
both cases. Thus the exhaust velocities are the same.

Cl = C2 = c = Isgo = 310 X 9.807 = 3040 m/sec

Case 1. The stage masses and sizes are the same, or

(m,)i = (m,)2 = nii

Equation 4-36 can be rewritten

gA„/c ^ (i/iv«j)(i/iv«)2 = {[m„ - (mp)i]/m„)[(m.)2 + mpi - (mpi)]

gS500/3040 ^ [( 45 QQ _ o.88mi)/4500][m; - mpi - 0.88mi]/(m; -f mpi) = 6.105
From the equation above m, = i(4500 — mpi).

These two equations have two unknowns: m; and mpi. The solution requires a quadratic
equation, and the value of the payload can be determined as mpi = 480 kg.

Case 2. The mass ratios for the two stages are equal, or

IVRi = ]VR2 = IVR = nto/nif

1/NR = [4500 - (mp)i]/4500 = [(m ;)2 + mpi - (mp) 2 ]/[(mj )2 -H mpi]

,Am/c _

4500 — 0.88(m,)i 0.12(m,)2 -f mpi
4500 (m;)2 + mpi

= ^ 5500/3040 ^

4500

4.7. FLIGHT VEHICLES 145

These two equations (and the first equation of the example makes a third) have three
variables: (m,) 2 , and m^\. The solution is possible and requires quadratic equations

and natural logarithms. The result is a payload of approximately 600 kg.

If a three-stage vehicle had been used in Example 4-3 instead of a two-
stage version, the payload would have been even larger. However, the theoretical
payload increase will only be about 8 or 10%. A fourth stage gives an even
smaller theoretical improvement; it would add only 3 to 5% to the payload. The
amount of potential performance improvement diminishes with each added stage.
Each additional stage means extra complications in an actual vehicle (such as a
reliable separation mechanism, an interstage structure, more propulsion systems,
joints or couplings in connecting pipes and cables, etc.), requires additional inert
mass (increasing the mass ratio IVR), and compromises the overall reliability.
Therefore, the minimum number of stages that will meet the payload and the Am
requirements is usually selected.

The flight paths taken by the vehicles in the two simplified cases of
Example 4-3 are different, since the time of flight and the acceleration histories
are different. One conclusion from this example applies to all multistage
rocket-propelled vehicles; for each mission there is an optimum number of
stages, an optimum distribution of the mass between the stages, and usually also
an optimum flight path for each design, where a key vehicle parameter such as
payload, velocity increment, or range is a maximum.

Launch Vehicles

Usually first or lowest stage, often called a booster stage, is the largest and it
requires the largest thrust and largest total impulse. Eor earth surface launch all
stages now use chemical propulsion to achieve the desired thrust-to-weight ratio.
These thrusts usually become smaller with each subsequent stage, also known
as upper stage or sustainer stage . The thrust magnitudes depend on the mass of
the vehicle, which in turn depends on the mass of the payload and the mission.
Typical actual configurations are shown by simple sketches in Fig. 4-14. There
is an optimum size and thrust value for each stage in a multistage vehicle and
the analysis to determine these optima can be quite complex.

Many launch vehicles with heavy payloads have one to six large strap-on
stages, also called zero stages or half stages. They augment the thrust of the
booster stage, which is started at about the same time. A schematic diagram
is shown in the third sketch of Fig. 4-14. Solid propellant strap-on stages are
common, such as the Atlas V shown in Fig. 1-13 or the Space Shuttle shown in
Fig. 1-14. They are usually smaller in size than the equivalent liquid propellant
strap-on (due to higher propellant density) and have less drag and usually a very
toxic exhaust. Liquid propellant strap-on stages are used in the Delta IV heavy
lift launch vehicle (see Fig. 1-12), in the first Soviet ICBM (intercontinental
ballistic missile, 1950s), and several Soviet/Russian space launch vehicles. They

146 FLIGHT PERFORMANCE

deliver higher specific impulse, which enhances vehicle performance, and require
propellant filling at the launch site.

There is a variety of existing launch vehicles. The smaller ones are for low
payloads and low orbits; the larger ones usually have more stages, are heav-
ier and more expensive, and have larger payloads or higher mission velocities.
The vehicle cost increases with the number of stages and the initial vehicle
launch mass. Once a particular launch vehicle has been proven to be reliable,
it is usually modified and uprated to allow improvements in its capability or
mission flexibility. Each of the stages of a space launch vehicle can have sev-
eral rocket engines, each for specific missions or maneuvers. The Space Shuttle
system has 67 different rockets which are shown schematically in Fig. 1-14.
In most cases each rocket engine is used for a specific maneuver, but in many
cases the same engine is used for more than one specific purpose; the small reac-
tion control thrusters in the Shuttle serve, for example, to give attitude control
(pitch, yaw, and roll) during orbit insertion and reentry, for counteracting internal
shifting of masses (astronaut movement, extendible arm), small trajectory correc-
tions, minor flight path adjustments, docking, and precise pointing of scientific
instruments.

The spacecraft is that part of a launch vehicle that carries the payload. It is the
only part of the vehiele that goes into orbit or deep space and some are designed
to return to earth. The final major space maneuver, such as orbit injection or
planetary landing, often requires a substantial velocity increment; the propulsion
system, which provides the force for this maneuver, may be integrated with the
spacecraft or it may be part of a discardable stage, just below the spacecraft.
Several of the maneuvers described in Section 4.5 can often be accomplished by
propulsion systems located in two different stages of a multistage vehicle. The
selection of the most desirable propulsion systems, and the decision of which
of the several propulsion systems will perform specific maneuvers, will depend
on optimizing performance, cost, reliability, schedule, and mission flexibility as
described in Chapter 19.

When a space vehicle is launched from the earth’s surface into an orbit, it flies
through three distinct trajectory phases. (1) Most are usually launched vertically
and then undergo a turning maneuver while under rocket power to point the flight
velocity vector into the desired direction. (2) The vehicle then follows a free-
flight (unpowered) ballistic trajectory (usually elliptical), up to its apex. Finally
(3) a satellite needs an extra push from a chemical rocket system to add enough
total impulse or energy to accelerate it to orbital velocity. This last maneuver
is also known as orbit insertion or sometimes as a kick maneuver. During the
initial powered flight the trajectory angle and the thrust cutoff velocity of the last
stage are adjusted by the guidance system to a velocity vector in space that will
allow the vehicle to reach the apogee of its elliptic path exactly at the desired
orbit altitude. As shown in Fig. 4-9, a multistage ballistic missile follows the
same two ascent flight phases mentioned above, but it then continues its elliptical
ballistic trajectory all the way down to the target.

4.7. FLIGHT VEHICLES 147

500

450

400

g 350

~ 300

Q-

250

200

150

100

^0 degrees inclination, equ

atorial drop pt

)int

^28.5 degree
s^^^38 de

S(ER)

‘grees (WFF)

/ 70 degree

\ /S

s (WR)

)0 degrees (W
/Sun-!

R), polar orbi
Synchronous

WR)

Assumes:

- 220 ft/sec

Velocity Reserv

System Kept on Launch V

- Direct Injection (No Dog-L
ER = Eastern Range

- WEE = Wallops Elight Eac

- WR = Western Range (Va

3hicle Side
3gs)

lity

ndenberg Air Fc

rce Base)

200 400 600 800 1000

Circular orbit altitude (km)

1200

1400

FIGURE 4-15. Decrease of payload with circular orbit altitude and orbit inclination for
early version of the Pegasus launch vehicle. This is an air-launched, relatively simple,
three-stage launch vehicle of 50 in. diameter driven by a solid propellant rocket motor in
each stage. (Courtesy Orbital Sciences Corporation.)

Historically successful launch vehicles have been modified, enlarged, and
improved in performance. The newer versions retain most of the old, proven, reli-
able components, materials, and subsystems. This reduces development effort and
cost. Upgrading a vehicle allows an increase in mission energy (more ambitious
mission) or payload or both. Typically, it is done by one or more of these types
of improvement: increasing the mass of propellant without an undue increase in
tank or case mass; uprating the thrust and strengthening the engine; more specific
impulse; or adding successively more or bigger strap-on boosters. It also usually
includes a strengthening of the structure to accept higher loads.

Figure 4-15 shows the effects of orbit inclination and altitude on the payload.
The inclination is the angle between the equatorial plane of the earth and the
trajectory. An equatorial orbit has zero inclination and a polar orbit has 90°
inclination. Since the earth’s rotation gives the vehicle an initial velocity, a launch
from the equator in an eastward direction will give the highest payload. For the
same orbit altitude other trajectory inclinations have a lower payload. For the
same inclination the payload decreases with orbit altitude, since more energy has
to be expended to overcome gravitational attraction.

148 FLIGHT PERFORMANCE

The Space Shuttle has its maximum payload when launched due east into
an orbit with 28.5° inclination from Kennedy Space Flight Center in Florida,
namely about 56,000 lb (or 25,455 kg) at a 100-nautical-mile (185-km) orbit
altitude. The payload decreases by about 100 lb (45.4 kg) for every nautical mile
increase in altitude. If the inclination is 57°, the payload diminishes to about
42,000 lb (or 19,090 kg). If launched in a southerly direction from Vandenberg
Air Force Base on the west coast in a 98° inclination into a circular, nearly polar
orbit, the payload will be only about 30,600 lb or 13,909 kg.

The dramatic decrease of payload with circular orbits of increasing altitude
and with different inclination is shown for the Pegasus, a relatively small, air-
launched space launch vehicle, in Fig. 4-15. The payload is a maximum when
launching from the earth equator in the east direction, that is, at 0° inclination.
The figure shows that a practical payload becomes too small for orbits higher
than about 1200 km. To lift heavier payloads and to go to higher orbits requires a
larger launch vehicle than this Pegasus. Figure 4-15 is based on the assumption
of a particular payload separation mechanism (38 in.) and a specific Am vehicle
velocity reserve (220 ft/sec) for items such as the normal changes in atmospheric
density (which can double the drag) or mass tolerances of the propulsion systems.
Similar curves can usually be provided by the makers of all launch vehicles.

4.8. MILITARY MISSILES

The majority of all rocket propulsion systems built today are for military pur-
poses. There is a large variety of missiles and military missions and therefore
many different propulsion systems. All are chemical propulsion systems. They
range from simple, small, unguided, fin-stabilized single-stage rocket projectiles
(used in air-to-surface missions and surface-to-surface bombardment) up to com-
plex, sophisticated, expensive, long-range, multistage ballistic missiles, which
are intended for faraway military or strategic targets. The term “surface” means
land surface (ground launch or ground target), ocean surface (ship launched),
or below the ocean surface (submarine launched). A tactical missile is used for
attacking or defending ground troops, nearby military or strategic installations,
military aircraft, or war missiles. The armed forces also use military satellites
for missions such as reconnaissance, early warning of impending attack, secure
communication, or navigation.

Strategic missiles with a range of 3000 km or more have been two- or
three-stage surface-to-surface rocket-propelled missiles. Early designs used liq-
uid propellant rocket engines and some are still in service in certain countries.
Beginning about 37 years ago, newer strategic missiles have used solid propellant
rocket motors in the United States and France. Both types usually also have a
liquid propellant RCS for accurately adjusting the final payload flight velocity
(in magnitude, direction, and position in space) at the cutoff of the propulsion
system of the last stage. A solid propellant RCS version also exists. The flight
analysis and ballistic trajectories of the long-range missiles are similar in many
ways to those described for launch vehicles in this chapter.

4.8. MILITARY MISSILES 149

Solid propellant rocket motors are preferred for most tactical missile missions,
because they allow simple logistics and can be launched quickly. If altitudes are
low and flight durations are long, such as with a cruise missile, an air-breathing jet
engine and a winged vehicle, which provides lift, will usually be more effective
than a long-duration rocket. However, a large solid propellant rocket motor may
still he needed as a booster to launch the cruise missile and bring it up to speed.

For each of the tactical missile applications, there is an optimum rocket propul-
sion system and almost all of them use solid propellant rocket motors. Liquid
propellant rocket engines have recently been used for the upper stages of two-
stage anti-aircraft missiles and ballistic defense missiles, because they can he
pulsed for different durations and can he randomly throttled. For each appli-
cation there is an optimum total impulse, an optimum thrust time profile, an
optimum nozzle configuration (single or multiple nozzles, with or without thrust
vector control, optimum area ratio), optimum chamber pressure, and a favored
solid propellant grain configuration. Low exhaust plume gas radiation emissions
in the visible, infrared, or ultraviolet spectrum and certain safety features (making
the system insensitive to energy stimuli) can be very important in some of the
tactical missile applications; these are discussed in Chapters 13 and 20.

Short-range, uncontrolled, unguided, single-stage rocket vehicles, such as mil-
itary rocket projectiles (ground and air launched) and rescue rockets, are usu-
ally quite simple in design. Their general equations of motion are derived in
Section 4.3, and a detailed analysis is given in Ref. 4-1.

Unguided military rocket-propelled missiles are today produced in larger num-
bers than any other category of rocket-propelled vehicles. The 2. 75-in. -diameter,
folding fin unguided solid propellant rocket missile has recently been produced
in the United States in quantities of almost 250,000 per year. Guided missiles for
anti-aircraft, antitank, or infantry support have been produced in annual quanti-
ties of hundreds and sometimes over a thousand. Table 1-6 lists several guided
missiles.

Because these rocket projectiles are essentially unguided missiles, the accuracy
of hitting a target depends on the initial aiming and the dispersion induced by
uneven drag, wind forces, oscillations, and misalignment of nozzles, body, and
fins. Deviations from the intended trajectory are amplified if the projectile is
moving at a low initial velocity, because the aerodynamic stability of a projectile
with fins is small at low flight speeds. When projectiles are launched from an
aircraft at a relatively high initial velocity, or when projectiles are given stability
by spinning them on their axis, their accuracy of reaching a target is increased
2- to 10-fold, compared to a simple fin-stabilized rocket launched from rest.

In guided air-to-air and surface-to-air rocket-propelled missiles the time of
flight to a given target, usually called the time to target f,, is an important flight
performance parameter. With the aid of Fig. 4- 16 it can be derived in a simplified
form by considering the distance traversed by the rocket (called the range) to
be the integrated area underneath the velocity-time curve. This simplification
assumes no drag, no gravity effect, horizontal flight, a relatively small distance
traversed during powered flight compared to the total range, and a linear increase

150 FLIGHT PERFORMANCE

Powered

j !

Probable actual curve '

allowing for drag and

non-linear change in

mass

tp Time f,

Maximum velocity
at burnout of
propellant

Actual velocity is
decreased by
drag

Velocity of

launching

aircraft

FIGURE 4-16. Simplified trajectory for an unguided, nonmaneuvering, air-launched
rocket projectile. Solid fine shows ideal flight velocity without drag or gravity and dashed
curve shows likely actual flight.

in velocity during powered flight:

S' -j- 2^p^p
Mo + Up

(4-37)

Here S is the flight vehicle’s range to the target and it is the integrated area
under the velocity -time curve. Also Up is the velocity increase of the rocket
during powered flight up to the time of burnout, tp is the time of rocket burning,
and uq is the initial velocity of the launching aircraft. For the same flight time the
range of the actual vehicle (dashed line) is less than for the ideal dragless vehicle.
For more accurate values, the velocity increase Up is given by Eq. 4-19. More
accurate values can also be obtained through a detailed step-by-step trajectory
analysis that considers the effects of drag and gravity.

In unguided air-to-air or air-to-surface rocket-powered projectiles the aiming
at the target is done largely by orienting the launching aircraft into the direction of
the target. A relatively simple solid propellant rocket motor is the most common
choice for the propulsion. In guided missiles, such as air-to-air, air-to-ground, or
ground-to-air, the flight path to the target has to be controlled and this can be
achieved by controlling aerodynamic control surfaces and/or propulsion systems,
which can be pulsed (repeated start and stop) and/or throttled to a lower thrust.
The guidance system and the target seeker system of a guided missile will sense
and track the flight path of a flying target, a computer will calculate a predicted
impact point, and the missile’s flight control will change the flight path of the
guided missile to achieve the impact with the intended target. The control system
will command the propulsion system to operate or fire selected liquid propellant
thrusters of an engine with multiple thrusters (or to selectively provide thrust
through multiple nozzles with hot-gas shut-off valves in solid motors). A sim-
ilar set of events can occur in a defensive ground-to-incoming-ballistic-missile

4.8. MILITARY MISSILES 151

scenario. It requires propulsion systems capable of pulsing or repeated starts, pos-
sibly with some throttling and side forces. Rocket engines with these capabilities
are discussed in Section 6.8.

In both the unguided projectile and the guided missile the hit probability
increases as the time to target t, is reduced. In one particular air-to-air combat
situation, the effectiveness of the rocket projectile varied approximately inversely
as the cube of the time to target. The best results (e.g., best hit probability) are
usually achieved when the time to target is as small as practically possible.

The analysis of the missile and propulsion configuration that gives the min-
imum time to target over all the likely flight scenarios can be complex. The
following rocket propulsion features and parameters will help to reduce the time
to target, but their effectiveness will depend on the specific mission, range, guid-
ance and control system, and particular flight conditions.

1. High initial thrust or high initial acceleration for the missile to quickly
reach a high-initial-powered flight velocity.

2. Application of additional lower thrust to counteract drag and gravity losses
and thus maintain a high flight velocity. This can be a single rocket propul-
sion system that has a short high initial thrust and a smaller (10 to 25%)
sustaining thrust of longer duration.

3. For higher supersonic flight speeds, a two-stage missile can be more effec-
tive. Here the first stage is dropped off after its propellant has been con-
sumed, thus reducing the inert mass of the next stage and improving its
mass ratio and thus its flight velocity increase.

4. If the target is highly maneuverable and if the closing velocity between
missile and target is large, it may be necessary not only to provide an
axial thrust but also to apply large side forces or side accelerations to a
defensive tactical missile. This can be accomplished either by aerodynamic
forces (lifting surfaces or flying at an angle of attack) or by multiple-nozzle
propulsion systems with variable or pulsing thrusts; the rocket engine then
would have an axial thruster and one or more side thrusters. The side
thrusters have to be so located that all the thrust forces are essentially
directed through the center of gravity of the vehicle in order to minimize
turning moments. The thrusters that provide the side accelerations have
also been called divert thrusters, since they divert the vehicle in a direction
normal to the axis of the vehicle.

5. Drag losses can be reduced if the missile has a large L/D ratio (or a small
cross-sectional area) and if the propellant density is high, allowing a smaller
missile volume. The drag forces can be high if the missile travels at low
altitude and high speed. A long and thin propulsion system geometry and
a high-density propellant will help to reduce drag.

A unique military application is rocket-assisted gun-launched projectiles for
attaining longer artillery ranges. Their small rocket motors withstand very high

152 FLIGHT PERFORMANCE

accelerations in the gun barrel (5000 to 10,000 go is typical). They have been in
production.

4.9. FLIGHT STABILITY

Stability of a vehicle is achieved when the vehicle does not rotate or oscil-
late in flight. Unstable flights are undesirable, because pitch or yaw oscillations
increase drag (flying at an angle of attack most of the time) and cause prob-
lems with instruments and sensors (target seekers, horizon scanners, sun sensors,
or radar). Instability often leads to tumbling (uncontrolled turning) of vehicles,
which causes missing of orbit insertion, missing targets, or sloshing of liquid
propellant in tanks.

Stability can be built in by proper design so that the flying vehicle will be
inherently stable, or stability can be obtained by appropriate controls, such as
the aerodynamic control surfaces on an airplane, a reaction control system, or
hinged multiple rocket nozzles.

Flight stability exists when the overturning moments (e.g., those due to a wind
gust, thrust misalignment, or wing misalignment) are smaller than the stabilizing
moments induced by thrust vector controls or by aerodynamic control surfaces.
When the destabilizing moments exceed the stabilizing moments about the center
of gravity, the vehicle turns or tumbles. In unguided vehicles, such as low-altitude
rocket projectiles, stability of flight in a rectilinear motion is achieved by giving
a large stability margin to the vehicle by using tail fins and by locating the center
of gravity ahead of the center of aerodynamic pressure. In a vehicle with an
active stability control system, a nearly neutral inherent stability is desired, so
that the applied control forces are small, thus requiring small control devices,
small RCS thrusters, small actuating mechanisms, and structural mass. Neutral
stability is achieved by locating aerodynamic surfaces and the mass distribution
of the components within the vehicle in such a manner that the center of gravity is
only slightly above the center of aerodynamic pressure. Because the aerodynamic
moments change with Mach number, the center of pressure does not necessarily
stay fixed during accelerating flight but shifts, usually along the vehicle axis.
The center of gravity also changes its position as propellant is consumed and the
vehicle mass decreases. Thus it is usually very difficult to achieve neutral missile
stability at all altitudes, speeds, and flight conditions.

Stability considerations affect rocket propulsion system design in several ways.
By careful nozzle design and careful installation it is possible to minimize thrust
misalignment and thus to minimize undesirable torques on the vehicle and the
reaction control propellant consumption. It is possible to exercise considerable
control over the travel of the center of gravity by judicious design. In liquid
propellant rockets, special design provisions, special tank shapes, and a careful
selection of tank location in the vehicle afford this possibility. By using nozzles
at the end of a blast tube, as shown in Fig. 15-6, it is possible to place the solid
propellant mass close to the vehicle’s center of gravity. Attitude control liquid

PROBLEMS 153

propellant engines with multiple thrusters have been used satisfactorily to obtain
control moments for turning vehicles in several ways, as described in Section 4.5
and in Chapter 6.

Unguided rocket projectiles and missiles are often given a roll or rotation
by inclined aerodynamic fins or inclined multiple rocket exhaust gas nozzles
to improve flight stability and accuracy. This is similar to the rotation given
to bullets by spiral-grooved barrels. This spin stability is achieved in part by
gyroscopic effects, where an inclination of the spin axis is resisted by torques. The
centrifugal effects cause problems in emptying liquid propellant tanks and extra
stresses on solid propellant grains. In some applications a low-speed roll is applied
not for spin stability but to assure that any effects of thrust vector deviations or
aerodynamic vehicle shape misalignments are minimized and canceled out.

PROBLEMS

1. For a vehicle in gravitationless space, determine the mass ratio necessary to boost the
vehicle velocity by (a) 1600 m/sec and (b) 3400 m/sec; the effective exhaust velocity
is 2000 m/sec. If the initial total vehicle mass is 4000 kg, what are the corresponding
propellant masses?

Answers: (a) 2204 kg.

2. Determine the burnout velocity and burnout altitude for a dragless projectile with the
following parameters for a simplified vertical trajectory: c = 2209 m/sec; nip/mo =
0.57; tp = 5.0 sec; and mq = 0; Hq = 0. Select a relatively small diameter missile with
UD of 10 and an average vehicle density of 1200 kg/m^.

3. Assume that this projectile had a drag coefficient essentially similar to the 0° curve in
Fig. 4-3 and redetermine the answers of Problem 3 and the approximate percentage
errors in Up and hp. Use a step-by-step or a numerical method.

4. A research space vehicle in gravity-free and drag-free outer space launches a smaller
spacecraft into a meteor shower region. The 2-kg sensitive instrument package of
this spacecraft (25 kg total mass) limits the maximum acceleration to no more than
50 m/sec^. It is launched by a solid propellant rocket motor (/j = 260 sec and f =
0.88). Assume instant start and stop of rocket motor.

(a) Determine the maximum allowable burn time, assuming steady constant propel-
lant mass flow;

(b) Determine the maximum velocity relative to the launch vehicle.

(c) Solve for (a) and (b) if half of the total impulse is delivered at the previous
propellant mass flow rate, with the other half at 20% of this mass flow rate.

5. For a satellite cruising in a circular orbit at an altitude of 500 km, determine the
period of revolution, the flight speed, and the energy expended to bring a unit mass
into this orbit.

Answers: 1.58 hr, 7613 m/sec, 33.5 MJ/kg.

6. A large ballistic rocket vehicle has the following characteristics: propellant mass flow
rate: 12 slugs/sec (1 slug = 32.2 Ibm = 14.6 kg); nozzle exit velocity: 7100 ft/sec;

154 FLIGHT PERFORMANCE

nozzle exit pressure: 5 psia (assume no separation); atmospheric pressure: 14.7 psia
(sea level); takeoff weight; 12.0 tons (Iton = 2000 Ihf); burning time; 50 sec; nozzle
exit area: 400 in.^ Determine (a) the sea-level thrust; (b) the sea-level effective
exhaust velocity; (c) the initial thrust-to-weight ratio; (d) the initial acceleration; (e)
the mass inverse ratio mo/ntf.

Answers: 81,320 Ibf; 6775 ft/sec; 3.38; 2.38go-

7. In Problem 5 compute the altitude and missile velocity at the time of power plant
cutoff, neglecting the drag of the atmosphere and assuming a simple vertical
trajectory.

8. A spherical satellite has 12 identical monopropellant thrust chambers for attitude
control with the following performance characteristics: thrust (each unit): 5 Ibf; 4
(steady state or more than 2 sec): 240 sec; 4 (pulsing duration 20 msec): 150 sec;
4 (pulsing duration 100 msec): 200 sec; satellite weight: 3500 Ibf; satellite diameter:
8 ft; satellite internal density distribution is essentially uniform; disturbing torques,
Y and Z axes: 0.00005 ft-lbf average; disturbing torque, for X axis: 0.001 ft-lbf
average; distance between thrust chamber axes: 8 ft; maximum allowable satellite
pointing position error: ±1°. Time interval between pulses is 0.030 sec.

(a) What would be the maximum and minimum vehicle angular drift per hour if no
correction torque were applied?

Answers: 0.466 and 0.093 rad/hr.

(b) What is the frequency of pulsing action (how often does an engine pair operate?)
at 20-msec, 100-msec, and 2-sec pulses in order to correct for angular drift?
Discuss which pulsing mode is best and which is impractical.

9 . For an ideal multistage launch vehicle with several stages in series, discuss the fol-
lowing: (a) the effect on the ideal mission velocity if the second and third stages are
not started immediately but are each allowed to coast for a short period after shutoff
and separation of the expended stage before rocket engine start of the next stage;
(b) the effect on the mission velocity if an engine malfunctions and delivers a few
percent less than the intended thrust but for a longer duration and essentially the full
total impulse of that stage.

10 . Given a cylindrically shaped space vehicle (D = Im, height is 0.7 m, average density
is 1.1 g/cm^) with a flat solar cell panel on an arm (mass of 32 kg, effective moment
arm is 1.5 m, effective average area facing normally toward sun is 0.6 m^) in a set
of essentially frictionless bearings and in a low circular orbit at 160 km altitude with
sunlight being received, on the average, about 54% of the time. The top surface of
the cylinder and the surface of the solar have solar cells. The satellite is maneuvered
so that these two surfaces are always perpendicular to the sun’s rays. The articulated
arm between the cylinder and the solar panel is small and its mass and force, due to
solar radiation pressure, are also small and can be neglected.

(a) Compute the maximum solar pressure-caused torque and the angular displacement
this would cause during 1 day if not corrected.

(b) Using the data from the atmospheric table in Appendix 2 and an arbitrary average
drag coefficient of 1.0 for both the body and the flat plate, compute the drag force
and torque.

SYMBOLS 155

(c) Using stored high-pressure air at 14 x 10® N/m^ initial pressure as the propellant
for attitude control, design an attitude control system to periodically correct for
these two disturbances etc.).

11 . Determine the payload for a single-stage vehicle in Example 4-3. Using the data
from this example compare it with the two-stage vehicle.

12 . An earth satellite is in an elliptical orbit with the perigee at 600 km altitude and an
eccentricity of e = 0.866. Determine the parameters of the new satellite trajectory,
if a rocket propulsion system is fired in the direction of flight giving an incremental
velocity of 200 m/sec when (a) fired at apogee, (b) fired at perigee, and (c) fired at
perigee, but in the opposite direction, reducing the velocity.

13 . A sounding rocket (75 kg mass, 0.25 m diameter) is speeding vertically upward at an
altitude of 5000 m and a velocity of 700 m/sec. What is the deceleration in multiples
of g due to gravity and drag? (Use Cd from Fig. 4-3 and use Appendix 2.)

14 . Derive Eq. 4-37; state all your assumptions.

SYMBOLS

kii

major axis of ellipse, m, or acceleration, m/sec^ (ft/sec^)

area, m^(ft^)

minor axis of ellipse, m

numerical value of drag integral

effective exhaust velocity, m/sec (ft/sec)

average effective exhaust velocity, m/sec

drag coefficient

lift coefficient

total derivative

drag force, N (Ibf)

eccentricity of ellipse, e = y/\ —

base of natural logarithm (2.71828)

energy, J

thrust force, N (Ibf)
final thrust, N

gravitational attraction force, N

initial thrust force, N

gravitational acceration, m/sec^

gravitational acceleration at sea level, 9.8066 m/sec^

average gravitational attraction, m/sec^

universal or Newton’s gravity constant,

6.6700 X 10^' m^/kg-sec^
altitude, m (ft)

altitude of rocket at power cutoff, m
specific impulse, sec
diffuse coefficient of reflectivity
specular coefficient of reflectivity
distance of moment arm, m

156 FLIGHT PERFORMANCE

nif

X, y

lift force, N (Ibf)

instantaneous vehicle mass, kg (Ibm)
final vehicle mass after rocket operation, kg
useful propellant mass, kg

initial vehicle launching mass, prior to rocket operation, kg

mass flow rate of propellant, kg/sec

angular moment of inertia, kg-m^

mass ratio of vehicle = trif /mo

number of stages

pressure, N/m^ or Pa (psi)

radius, m, or distance between the centers of two attracting
masses, m

instantaneous radius from vehicle to center of earth, m
effective mean earth radius, 6.3742 x 10^ m
range, m
time, sec

time from launching to power cutoff or time from propulsion

start to thrust termination, sec

time to target, sec

torque, N-m (ft-lbf)

vehicle flight velocity, m/sec (ft/sec)

orbital velocity at apogee, m/sec

velocity at power cutoff, m/sec, or orbital velocity at perigee,
m/sec

initial or launching velocity, m/sec
escape velocity, m/sec
weight, N (in some problems, Ibf)
any points on an elliptical orbit

Greek Letters

a angle of attack, deg or rad, or angular acceleration, angle/sec^

i; propellant mass fraction (f = mp/mo)

9 angle between flight direction and horizontal, or angle of incident

radiation, deg or rad

/X gravity constant for earth, 3.98600 x lO'^ m^/sec^

p mass density, kg/m^

T period of revolution of satellite, sec

t]/ angle of thrust direction with horizontal

CO angular speed, deg/sec (rad/sec)

Subscripts

escape condition

final condition at rocket thrust termination
initial condition

REFERENCES 157

max

maximum

condition at power cutoff or propulsion termination

payload

satellite

zenith

initial condition or takeoff condition

REFERENCES

4-1. J. B. Rosser, R. R. Newton, and G. L. Gross, Mathematical Theory of Rocket
Flight, McGraw-Hill Book Company, New York, 1947; or F. R. Gantmakher and
L. M. Levin, The Flight of Uncontrolled Rockets, Macmillan, New York, 1964.

4-2. R. S. Wolf, “Development of a Handbook for Astrobee F (Sounding rocket) Flight
Performance Predictions,” Journal of Spacecraft and Rockets, Vol. 24, No. 1,
January -February 1987, pp. 5-6.

4-3. Orbital Flight Handbook, NASA SP33, 1963, Part 1: Basic Techniques and Data.
Part 2: Mission Sequencing Problems. Part 3: Requirements.

4-4. V. A. Chobotov (Ed) Orbital Mechanics , 3rd ed.. Educational Series, AIAA,
Reston, VA., 2002; and T. Logsdon, Orbital Mechanics and Application, John
Wiley & Sons, New York, October 1997.

4-5. J. W. Cornelisse, H. F. R. Schoyer, and K. F. Wakker, Rocket Propulsion and
Space Flight Dynamics , Pitman Publishing, Londn, 1979.

4-6. J. P. Vinti, G. J. Der, and L. Bonavito, Orbits and Celestial Mechanics, Vol. 177
of Progress in Aeronautics and Astronautics Series, AIAA, Reston, VA, 1998,
409 pages.

4-7. W. Hohmann, Die Erreichbarkeit der Himmelskbrper (Accessibility of Celestial
Bodies), Oldenburg, Munich, 1925.

4-8. W. J. Larson and J. R. Wertz, Space Mission Analysis and Design, 3rd ed., pub-
lished jointly by Microcosm, Inc. and Kluwer Academic Press, 1999.

4-9. 1. 1. Pocha, An Introduction to Mission Design for Geostationary Satellites, Kluwer
Academic Publishers, Hingham, MA, 1987, 222 pages.

4-10. M. H. Kaplan, Orbital Spacecraft Dynamics and Control, John Wiley & Sons,
New York, 1976.

4-11. R. W. Humble, G. N. Henry, and W. J. Larson, Space Propulsion Analysis and
Design, McGraw-Hill, New York, 1995, 748 pages.

4-12. J. P. Vinti, G. J. Der, and L. Bonavito, Orbit and Celestial Mechanics, Vol. 177
of Progress in Astronautics and Aeronautics Series, AIAA, Reston, VA, 1998,
409 pages.

4-13. C-C. G. Chao, Applied Orbit Perturbations and Maintenance, Aerospace Press,
2005, 264 pages.

4-14. “Spacecraft Aerodynamic Torques,” NASA SP 8058, January 1971 (N 71-25935).

4-15. “Spacecraft Radiation Torques,” NASA SP 8027, October 1969 (N 71-24312).

4-16. “Spacecraft Gravitational Torques,” NASA SP 8024, May 1964 (N 70-23418).

4-17. “Spacecraft Magnetic Torques,” NASA SP 8018, March 1969 (N 69-30339).

CHAPTER 5

CHEMICAL ROCKET PROPELLANT
PERFORMANCE ANALYSIS

In Chapter 3, simplified one-dimensional performance relations were developed.
They require a knowledge of the composition of the hot rocket gas and
the properties of the propellant reaction products, such as their combustion
temperature T\, average molecular mass Tl, and the specific heat ratio or
the enthalpy change {hi — /j 2 )- This chapter discusses several approaches to
determine these theoretical thermochemical properties for a given composition
or mixture of propellant, chamber pressure, nozzle shape, and nozzle exit
pressure. This then allows the determination of performance parameters, such
as theoretical specific impulse or exhaust velocity for chemical rockets.

By knowing the calculated gas temperature, pressure, and gas composition
it is possible to calculate other gas properties. This knowledge also allows
the analysis and selection of materials for chamber and nozzle structures.
Heat transfer analyses require the determination of the specific heats, thermal
conductivity, and specific heat ratio for the gas mixture. The calculated exhaust
gas composition forms the basis for estimating environmental effects, such as
the potential spreading of a toxic cloud near a launch site, as discussed in
Chapter 21. The exhaust gas parameters also form the basis for the analysis of
exhaust plumes (Chapter 20) or flames external to the nozzle.

With the advent of digital computers it has been possible to solve the set of
equations involving mass balance, energy balance, together with thermodynamic
and chemical equilibria of complex systems with a variety of propellant ingre-
dients. This chapter is intended to introduce the background to this theoretical
analysis, so the reader can understand the thermodynamic and chemical basis
of the several computer programs that are in use today. This chapter does not
describe any specific computer analysis programs. However, it discusses which

158

5.1. BACKGROUND AND FUNDAMENTALS 159

of the physical phenomena or chemical reactions can or cannot be adequately
simulated by computer analysis.

The reader is referred to Refs. 5-1 to 5-5 for general chemical and thermo-
dynamic background and principles. For a detailed description of the properties
of each of the possible reactant and reaction products, see Refs. 5-6 to 5-12.

All of these theoretical analyses are only approximations of what really
happens in rocket combustion and nozzle flow, since they all require some
simplifying assumptions. As more of the different phenomena are understood
and mathematically simulated, the analysis approach and the computer imple-
mentation become more realistic but also more complex. The 11 assumptions
made in Section 3.1 for an ideal rocket are valid here also but only for a
quasi-one-dimensional flows. However, more sophisticated analyses can make
one or more of these assumptions unnecessary. The analysis is usually divided
into two somewhat separate sets of calculations:

1. The combustion process is the first part. It usually occurs in the combus-
tion chamber at essentially constant chamber pressure (isobaric) and the
resulting gases follow Dalton’s law, which is discussed in this chapter.
The chemical reactions or the combustions occur very rapidly. The cham-
ber volume is assumed to be large enough and the residence time in the
chamber long enough for attaining chemical equilibrium in the chamber.

2. The nozzle gas expansion process constitutes the second set of calculations.
The fully reacted, equilibrated gas combustion products enter the nozzle
and undergo an adiabatic expansion in the nozzle. The entropy remains
constant during a reversible (isentropic) nozzle gas expansion, but in real
nozzles it increases slightly.

The principal chemical reactions occur inside the combustion chamber of a
liquid propellant rocket engine or inside the grain cavity of a solid propellant
rocket motor, usually within a short distance from the burning surface. These
chamber combustion analyses are discussed further in Chapters 9 and 14. How-
ever, some chemical reactions also occur in the nozzle as the gases expand;
the composition of the reaction products can therefore change in the nozzle, as
described in this chapter. A further set of chemical reactions can occur in the
exhaust plume outside the nozzle, as described in Chapter 20; many of the same
basic thermochemical analysis approaches described in this chapter also apply to
exhaust plumes.

5.1. BACKGROUND AND FUNDAMENTALS

The analytical description of chemical reaction or combustion of one or more
fuels with one or more oxidizing reactants is the basis of chemical rocket propul-
sion. The heat liberated in this reaction transforms the propellants (reaetants) into

160 CHEMICAL ROCKET PROPELLANT PEREORMANCE ANALYSIS

hot gaseous products, which in turn are thermodynamically expanded in a nozzle
to produce thrust.

The chemical reactants or propellants can initially be either liquid or solid
and occasionally also gaseous. The reaction products are usually gaseous, but
with some propellants one or more reactant species remain in the solid or liquid
phase. For example, with aluminized solid propellants, the chamber reaction gases
contain liquid aluminum oxide and the colder gases in the nozzle exhaust contain
solid, condensed aluminum oxide particles. For some of the chemical species,
therefore, the analysis must consider as many as all three phases and the energy
changes for the phase transitions must be included. When the amount of solid or
liquid in the exhaust is small and the particles are small, to assume a perfect gas
introduces only small errors.

It is necessary to accurately know the chemical composition of the propel-
lants and their relative proportion. In liquid propellant this means the mixture
ratio and the major propellant impurities; in gelled or slurried liquid propellants
it also includes suspended or dissolved solid materials; and in solid propellants
it means all the ingredients, their proportions and impurities, and phase (some
ingredients, such as plasticizers, can be in a liquid state).

Dalton’s law applies to the gas resulting from the combustion. It states that
a mixture of gases at equilibrium exerts a pressure that is the sum of the par-
tial pressures of the individual gases, all at a common volume and a common
temperature. The subscripts a,b,c, and so on refer to individual gas constituents:

The perfect gas equation pV = RT applies very closely to high-temperature
gases. Here V is the specific volume or the volume per unit mass of gas mixture,
and the gas constant R for the mixture is obtained by dividing the universal
gas constant R' (8314.3 J/kg-mol-K) by the average moleeular mass Wl (often
erroneously called the molecular weight) of the gas mixture. Using Dalton’s law,
Eq. 5-1 can be rewritten

The volumetric proportions of gas species in a gas mixture are determined from
the molar concentration or molar fractions , iij, expressed as kg-mol for a partic-
ular species j per kg of mixture. If n is the total number of kg-mol of all species
per kilogram of uniform gas mixture, then the mole fraction Xj

P = Pa + Pb + Pc

T = Ta = Tb = Tc=-

(5-1)

(5-2)

p = RaTIVa + RbT/Vb + RcT/Vc + ■■■= R'TUmVr^^) (5-3)

]=m

Xj = — since
n

(5-4)

5.1. BACKGROUND AND FUNDAMENTALS 161

where nj is the kg-mol of species j per kilogram of mixture, m is the number
of different gaseous species present in the equilibrium combustion gas products.
The effective average molecular mass 971 of a gas mixture is then

971 =

E m

j=l "7

(5-5)

There are n possible species which enter into the relationship and of these
only m are gases, so n — m represents the number of condensed species. The
molar specific heat for a gas mixture at constant pressure Cp can be determined
from the individual gas molar fractions Uj and their molar specific heats as shown
by Eq. 5-6. The specific heat ratio k of the mixture follows from Eq. 5-7:

(6^»)mix —

E m

,/=i «;■
(6p)mix

(Cp)mix - R'

(5-6)

(5-7)

When a chemical reaction goes to completion, that is, all of the reactants are
consumed and transformed into reaction products, the reactants are in stoichio-
metric proportions. For example, consider this reaction:

H2 + ^02 -> H 2 O (5-8)

All the hydrogen and oxygen are fully consumed to form the single
product — water vapor — without any reactant residue of either hydrogen or
oxygen. In this ease it requires 1 mol of the H 2 and | mole of the O 2 to obtain
1 mol of H 2 O. On a mass basis this stoichiometric mixture requires half of
32.0 kg of O 2 and 2 kg of H 2 , which are in the stoichiometric mixture mass
ratio of 8:1. The release of energy per unit mass of propellant mixture and the
combustion temperature are highest at or near the stoichiometric mixture.

Rocket propulsion systems usually do not operate with the proportion of their
oxidizer and fuel in the stoichiometric mixture ratio. Instead, they usually operate
fuel rich because this allows lightweight molecules such as hydrogen to remain
unreacted; this reduces the average molecular mass of the reaction products,
which in turn increases the specific impulse (see Eq. 3-16). For rockets using
H 2 and O 2 propellants the best operating mixture mass ratio for high-performance
rocket engines is typically between 4.5 and 6.0, not at the stoichiometric value
of 8.0, because the drop in combustion temperature is small and there is more
H 2 gas (low molecular mass) in the exhaust.

Equation 5-8 is a reversible chemical reaction; by adding energy to the H 2 O
the reaction can be made to go backward to create H 2 and O 2 and the arrow
in the equation would be reversed. The decompositions of solid propellants into
reaction product gases are irreversible chemical reactions, as is the reaction of

162 CHEMICAL ROCKET PROPELLANT PEREORMANCE ANALYSIS

liquid propellants burning to create gases. However, reactions among combustion
product gases are usually reversible.

Chemical equilibrium exists in reversible chemical reactions when the rate of
forming products is exactly equal to the reverse reaction of forming reactants from
the products. Once this equilibrium is reached, no further changes in concentration
can take place. In Equation 5-8 all three gases would be present and their relative
proportions would depend on the pressure, temperature, and initial mixture.

The heat of formation AfH^ is the energy released (or absorbed), or the
value of enthalpy change, when 1 mol of a chemical compound is formed from
its constituent atoms or elements at 1 bar (100,000 Pa) and isothermally at
298.15 K or 25°C. The A implies that it is an energy change. The subscript
/ refers to formation and the superscript 0 means that each product or reactant
substance is at its thermodynamic standard state and at the reference pressure and
temperature. By convention, the heat of formation of the gaseous elements (e.g.,
H 2 , 02 , Ar, Xe, etc.) is set to zero at these standard conditions of temperature and
pressure. Typical values of AjH^ and other properties are given in Table 5-1
for selected species. When heat is absorbed in the formation of a product, then
AfH^ has a positive value. Earlier analyses have been made with the standard
temperature at other values, such as 273.15 K and a slightly higher standard
reference pressure of 1 atm (101, 325 Pa).

The heat of reaction ArH^ is the energy released or absorbed when products
are formed from its reactants at standard reference conditions, namely at 1 bar and
25°C. The heat of reaction can be negative or positive, depending on whether the
reaction is exothermic or endothermic. The heat of reaction at other temperatures
or pressures has to be corrected in accordance with the change in enthalpy. When
a species changes from one state to another (e.g., liquid becomes gas or vice
versa), it may lose or gain energy. In most rocket propulsion the heat of reaction
is determined for a constant-pressure combustion process. In general the heat of
reaction can be determined from sums of the heats of formation of the products
and the reactants, namely

ArH^ = ^ [Uj (A/H‘');]p„duct.s - Y. [”/■ (A/-f^‘’);]reactants (5-9)

Here nj is the molar fraction of each particular species j. In a typical rocket
propellant there are a number of different chemical reactions going on simul-
taneously; Eq. 5-9 provides the heat of reaction for all of these simultaneous
reactions. For data on heats of formation and heats of reaction, see Refs. 5-7 to
5-13.

Various thermodynamic criteria that represent the necessary and sufficient con-
ditions for an equilibrium to be stable were first advanced by J. W. Gibbs early
in the 20th century; they are based on minimizing the free energy. The Gibbs
free energy G (often called the chemical potential) is a convenient derived func-
tion or property of the state of a chemical material describing its thermodynamic
potential and is directly related to the internal energy U , the pressure p, molar
volume V, enthalpy h, temperature T, and entropy S. Eor a single species j the

5.1. BACKGROUND AND FUNDAMENTALS 163

TABLE 5-1. Chemical Thermodynamic Properties of Selected Substances at 298.15 K
(25°C) and 0.1 MPa (1 bar)

Substance

Phase"

Molar Mass
(g/mol)

AfH°

(kj/mol)

A/G®

(kJ/mol)

S°

(J/mol-K)

A1 (crystal)

29.9815

28.275

24.204

AI 2 O 3

101.9612

-1620.567

-1532.025

67.298

79.015

C (graphite)

12.011

5.740

8.517

CH 4

16.0476

-74.873

-50.768

186.251

35.639

28.0106

-110.527

-137.163

197.653

29.142

CO 2

44.010

-393.522

-394.389

213.795

37.129

H 2

2.01583

130.680

28.836

HCl

36.4610

-92.312

-95.300

186.901

29.136

20.0063

-272.546

-274.646

172.780

29.138

H 2 O

18.01528

-285.830

-237.141

69.950

75.351

H 2 O

18.01528

-241.826

-228.582

188.834

33.590

N 2 H 4

32.0451

-H50.626

149.440

121.544

98.840

N 2 H 4

32.0451

-b95.353

-b 159.232

238.719

50.813

NH 4 CIO 4

117.485

-295.767

-88.607

184.180

128.072

CIF 5

130.4450

-238.488

-146.725

310.739

97.165

CIF 3

92.442

-158.866

-118.877

281.600

63.845

N 2 O 4

92.011

-19.564

+97.521

209.198

142.509

N 2 O 4

92.011

9.079

97.787

304.376

77.256

NO 2

46.0055

33.095

51.258

240.034

36.974

HNO 3

63.0128

-134.306

-73.941

266.400

53.326

N 2

28.0134

191.609

29.125

O 2

31.9988

205.147

29.376

NH 3

17.0305

-45.898

-16.367

192.774

35.652

“s = solid, 1 = liquid, g = gas. Several species are listed twice, as a liquid and as a gas; the difference
is due to evaporation or condensation.

The molar mass can be in g/g-mol or kg/kg-mol and Cp can be in J/g-mol-K or kJ/kg-mol-K.
Source: Refs. 5-8 and 5-9.

free energy is defined as G,-; it can be determined for specific thermodynamic
conditions, for mixtures of gas as well as an individual gas species:

G^U +pV -TS =h-TS (5-10)

For most materials used as rocket propellant, the free energy has been deter-
mined and tabulated as a function of temperature. It can be corrected for pressure.
G;’s units are J/kg-mol. For a series of different species the mixture free energy
G is

j=i

(5-11)

164 CHEMICAL ROCKET PROPELLANT PEREORMANCE ANALYSIS

The free energy is a function of temperature and pressure. It is another property
of a material, just like enthalpy or density; only two such independent parameters
are required to characterize a gas condition. The free energy may he thought of
as the tendency or driving force for a chemical material to enter into a chemical
(or physical) change. Although it cannot he measured directly, differences in
chemical potential can he measured. When the chemical potential of the reactants
is higher than that of the likely products, a chemical reaction can occur and the
chemical composition can change. The change in free energy AG for reactions
at constant temperature and pressure is the chemical potential of the products
less that of the reactants:

m n

AG = Y1 [«;(A/G‘’);]products - ^ [«; (A/ G°); ] reactants (5-12)

j=i i=i

Here the superscript m gives the number of gas species in the combustion
products, the superscript n gives the number of gas species in the reactants, and
the AG represents the maximum energy that can be “freed” to do work on an
“open” system where mass enters and leaves the system. At equilibrium the free
energy is a minimum; at its minimum a small change in mixture fractions causes
almost no change in AG, and the free energies of the products and the reactants
are essentially equal. Then

dAGIdn = 0

(5-13)

and a curve of molar concentration n versus AG would have a minimum.

If reacting propellants are liquid or solid materials, energy will be needed to
change phase, vaporize them, or break them down into other gaseous species. This
energy has to be subtracted from the heat or the energy available to heat the gases
from the reference temperature to the combustion temperature. Therefore, the val-
ues of AH^ and AG° for liquid and solid species are different from those of the
same species in a gaseous state. The standard /ree energy of formation AfG^ is
the increment in free energy associated with the reaction of forming a given com-
pound or species from its elements at their reference state. Table 5-2 gives values
of AfH^ and AfG^ and other properties of carbon monoxide as a function of tem-
perature. Similar data for other species can be obtained from Refs. 5-7 and 5-13.
The entropy is another thermodynamic property of matter that is relative, which
means that it is determined as a change in entropy. In the analysis of isentropic
nozzle flow, it is assumed that the entropy remains constant. It is defined as

dU p dV dT dp

dS = + = C„ R —

T T T p

and the corresponding integral is

5 - 5 °

In— - R \n—
To Po

(5-14)

(5-15)

5.2. ANALYSIS OF CHAMBER OR MOTOR CASE CONDITIONS 1 65

TABLE 5-2. Variation of Thermochemical Data with Temperature for Carbon
Monoxide (CO) as an Ideal Gas

Temp.

(K)

(J/mol-K)

//® -//®(D
(kj/mol)

AfH°

(kJ/mol)

A/G®

(kJ/mol)

-8.671

-113.805

298.15

29.142

197.653

110.527

-137.163

500

29.794

212.831

5.931

-110.003

-155.414

1000

33.183

234.538

21.690

-111.983

-200.275

1500

35.217

248.426

38.850

-115.229

-243.740

2000

36.250

258.714

56.744

-118.896

-286.034

2500

36.838

266.854

74.985

-122.994

-327.356

3000

37.217

273.605

93.504

-127.457

-367.816

3500

37.493

279.364

112.185

-132.313

-407.497

4000

37.715

284.386

130.989

-137.537

-446.457

Source: Refs. 5-8 and 5-9.

where the zero applies to the reference state. In an isentropic process, entropy is
constant. For a mixture the entropy is

S =Y^Sfnj (5-16)

j=i

Here entropy Sj is in J/kg-mol-K. The entropy for each gaseous species is

Sj = (S^)j - R In^ -Rlnp (5-17)

For solid and liquid species the last two terms are zero. Here (S'®) refers to the
standard state entropy at temperature T. Typical values for entropy are listed in
Tables 5-1 and 5-2.

5.2. ANALYSIS OF CHAMBER OR MOTOR CASE CONDITIONS

The objectives here are to determine the theoretical combustion temperature
and the theoretical composition of the resulting reaction products, which in turn
will allow the determination of the physical properties of the combustion gases
{Cp,k,p, or other). Before we can make this analysis, some basic data (e.g.,
propellants, their ingredients, desired chamber pressure, or all likely reaction
products) have to be known or postulated. Although the combustion process
really consists of a series of different chemical reactions that occur almost simul-
taneously and includes the breakdown of chemical compounds into intermediate
and subsequently into final products, the analysis is only concerned with the ini-
tial and final conditions, before and after combustion. We will mention several

166 CHEMICAL ROCKET PROPELLANT PEREORMANCE ANALYSIS

approaches to the analysis of chamber conditions. In this section we will first
give some definitions of key terms and explain some concepts and principles.

The first principle concerns the conservation of energy. The heat created by the
combustion is equal to the heat necessary to raise the resulting gases adiabatically
to their final combustion temperature. The heat of reaction of the combustion
ArH has to equal the enthalpy change AH of the reaction product gases. The
energy balance can be thought of as a two-step process. The chemical reaction
occurs instantaneously but isothermally at the reference temperature, and the
resulting energy release then heats the gases from this reference temperature to
the final combustion temperature. The heat of reaction, Equation 5-9, becomes

Here Ah is the increase in enthalpy for each species multiplied by its molar
fraction, and Cp is the molar specific heat at constant pressure.

The second principle is the conservation of mass. The mass of any of the
atomic species present in the reactants before the chemical reaction must be
equal to the mass of the same species in the products. This can be illustrated by
a more general case of the reaction of Equation 5-8. In this case the reactants
are not in stoichiometric proportion.

In the combustion of hydrogen with oxygen it is possible to form six products:
water, hydrogen, oxygen, hydroxyl, atomic oxygen, and atomic hydrogen. In this
case all the reactants and products are gaseous. Theoretically, there could be two
additional products: ozone O 3 and hydrogen peroxide H 2 O 2 ; however, these are
unstable materials that do not readily exist at high temperature, and they can be
ignored. In chemical notation this can be stated by

aH 2 + b 02 — > «h2oH20 -|- mh2H2 + no.fi 2 + wqO -H «hH -|- «ohOH (5-19)

The left side shows the condition before and the right side the condition after
the reaction. Since H 2 and O 2 can be found on both sides, it means that not all
of these species are consumed and a portion, namely mh 2 « 02 > will remain
unreacted. With chemical equilibrium at a particular temperature and pressure
the molar concentrations on the right side will remain fixed. Here a, b, n^^o,
nii^,no^,no,ni{, and non are the respective molar quantities of these substances
before and after the reaction, and they can be expressed in kg-mol per kilogram
of propellant reactants or reaction products. The initial proportions of a and b are
usually known. The number of kg-mol per kilogram of mixture of each element
can be established from this initial mix of oxidizer and fuel ingredients. Eor the
hydrogen -oxygen relation above, the mass balances would be

(5-18)

for hydrogen: 2a = 2wh20 + 2«h2 + «h + «oh

for oxygen: 2b = «h 20 + 2«02 + «o + «oh

(5-20)

5.2. ANALYSIS OF CHAMBER OR MOTOR CASE CONDITIONS 1 67

The mass balance of Eq. 5-20 provides two more equations for this reac-
tion (one for each atomic species) in addition to the energy balance equation.
There are six unknown product percentages and an unknown combustion or
equilibrium temperature. However, three equations provide a solution for only
three unknowns, say the combustion temperature and the molar fractions of two
of the species. If, for example, it is known that the initial mass mixture ratio of
b/a is fuel rich, so that the combustion temperature will be relatively low, the
percentage of remaining O 2 and the percentage of the dissociation products (O,
H, and OH) would all be very low and can be neglected. Thus no,nn,non, and
«02 set to be zero. The solution requires knowledge of the enthalpy change
of each of the species, and that information can be obtained from existing tables,
such as Table 5-2 or Refs. 5-8 and 5-9.

In more general form, the mass for any given element must be the same before
and after the reaction. The number of kg-mol of a given element per kilogram
of reactants and product is equal, or their difference is zero. For any one atomic
species, such as the H or the O in Eq. 5-20,

_;=i

products

_;=i

= 0

(5-21)

J propellants

Here the atomic coefficients ajj are the number of kilogram atoms of element
i per kg-mol of species j, and m and n are as defined above. The average
molecular mass of the products from Eq. 5-5 would be

^ _ 2«h2 + 32«02 + I8WH2O + 16 «o + + 17«oh

, + no2 + ”H20 + «o + «h + «OH

(5-22)

The approach used in Ref. 5-13 is commonly used today for thermochemical
analysis. It relies on the minimization of the Gibbs free energy and on mass
balance and energy balance equations. As was explained in Eq. 5-12, the change
in the Gibbs free energy function is zero at equilibrium (AG = 0): the chemical
potential of the gaseous propellants has to equal that of the gaseous reaction
products, which is Eq. 5-12:

AG = L {Hj a ) products AGj)

reactants — 0 (5-23)

To assist in solving this equation a Lagrangian multiplier or a factor of the
degree of the completion of the reaction is often used. An alternative method
for solving for the gas composition, temperature, and gas properties is to use the
energy balance (Eq. 5-18) together with several mass balances (Eq. 5-21) and
equilibrium constant relationships.

After assuming a chamber pressure and setting up the energy balance, mass
balances, and equilibrium relations, one method of solving all the equations is to
estimate a combustion temperature and then solve for the various values of

168 CHEMICAL ROCKET PROPELLANT PEREORMANCE ANALYSIS

Then a balance has to be achieved between the heat of reaction A,.H^ and the
heat absorbed by the gases, Hj — H^, to go from the reference temperature to
the combustion temperature. If they do not balance, another value of the com-
bustion temperature is chosen until there is convergence and the energy balances.

The energy release ejficiency , sometimes called the combustion ejficiency , can
now be defined as the ratio of the actual change in enthalpy per unit propellant
mixture to the calculated change in enthalpy necessary to transform the propel-
lants from the initial conditions to the products at the chamber temperature and
pressure. The actual enthalpy change can be evaluated if the initial propellant con-
dition and the actual composition and the temperature of the combustion gases
are measured. Experimental measurements of combustion temperature and gas
composition are difficult to perform accurately, and the combustion efficiency is
therefore actually evaluated only in rare instances. The combustion efficiency in
liquid propellant rocket thmst chambers depends on the method of injection and
mixing and increases with increased combustion temperature. In solid propellants
the combustion efficiency is a function of the grain design, the propellant, and the
degree of mixing between the several solid constituents. Actual measurements
on well-designed rocket propulsion systems indicate efficiency values of 94 to
99%. These high values indicate that the combustion is essentially complete, that
very little, if any, unreacted propellant remains, and that chemical equilibrium is
indeed established.

The number of compounds or species in the exhaust can be 50 or more with
solid propellants or with liquid propellants that have certain additives. The num-
ber of nearly simultaneous chemical reactions that have to be considered can
easily exceed 150. Fortunately, many of these chemical species are present only
in very small amounts and can usually be neglected.

Example 5-1. Hydrogen peroxide is used both as a monopropellant and as an oxidizer
in bipropellant systems. It is stored in liquid form and available in various degrees of
dilution with liquid water. For rocket applications, concentrations (70 to 98'*'%), known
as high-test peroxide (HTP) are used. For a monopropellant application, calculate the
adiabatic flame temperature as a function of water content based on an initial mixture
temperature of 298.15 K (the standard condition).

SOLUTION. In this usage, hydrogen peroxide dissociates while passing through a catalyst
and releases energy which goes to increase the propellant temperature in the absence of
any heat transfer losses. But some of this heat will be required to evaporate the diluent
water. The mass balance, Eq. 5-21, is satisfied by 2 mol of hydrogen peroxide in n moles
of liquid water, producing n + 2 mol of water vapor plus 1 mol of oxygen gas. Since the
reaction goes to completion, no equilibrium constant is needed:

2H202(/) -f «H20(/) ^ (n -h 2)H20(g) -f 02(g)

The symbols (/) and (g) refer to the liquid state and the gaseous state, respectively.
The heats of formation from the standard state, A///° and molar specific heats Cp are
shown below (see Table 5-1 and other common sources such as the NIST Chemistry

5.2. ANALYSIS OF CHAMBER OR MOTOR CASE CONDITIONS 1 69

web-book, http://webbook.nist.gov/chemistry/). For these calculations the heat of mixing
may he ignored.

Species

AfH° (kJ/kg-mol)

Cp (J/kg-mol-k)

9Jl (kg/kg-mol)

H202(/)

-187.69

34.015

H20(/)

-285.83

18.015

H20(g)

-241.83

0.03359

18.015

02(g)

0.02938

31.999

The energy balance, Eq. 5-9, for 2 mol of decomposing hydrogen peroxide becomes

ArH^ = [nAfH°]H20 — [nAfH°]H202

= 2 X (-241.83) - 2 X (-187.69)

= -108.28 kJ

The reaction is exothermic but as stated some of this energy is used up in vaporizing the
diluent liquid water, namely 285.83 — 241.83 = 44.0 kj/kg-mol (at standard conditions).
The net available heat release thus becomes 108.28 — 44. On (kJ). In order to calculate
the adiabatic temperature, we assume ideal-gas heating at constant pressure, Eq. 5-18
(values for the molar specific heats are from Table 5-1 and are taken as constant):

j (wH 2 oCpH 20 + MO 2 CPO 2 ) dT = [(2 + n) C;,H 20 + Cp 02 ] AT = 108.28 — 44. On

It will be more convenient to give results in terms of a mass fraction z of the diluent
water in the original mixture, and for this the molecular masses need to be inserted (also
shown in Table 5-1):

z = niH2o/('«H202 + 'WH 20 ) = n 9J£h2o/(2 971h202 + ” ®Th2o)

= 18.015n/(2 X 34.015 -b 18.015n)

Now solve for n

n =3.78z/(l-z)

and substitute it in the relation for the temperature. The resulting value (the adiabatic
temperature. Tad) may then he plotted in terms of the mass fraction z with the initial
temperature as given:

Tad = [108.28 - 44 X 3.78z/(l - z)]/[0.03359 x (2 -b 3.78z/(l - z))

-b 0.02838] -b 298. 15d

The figure also displays the values of c* (where T\ = Tad), 971, and k, which are
calculated according to Eqs. 3-32 and 5-5 and 5-7.

170

CHEMICAL ROCKET PROPELLANT PEREORMANCE ANALYSIS

5.3. ANALYSIS OF NOZZLE EXPANSION PROCESSES

There are several methods for analyzing the nozzle flow, depending on the
assumptions made for chemical equilibrium, nozzle expansion, particulates, or
energy losses. Several are outlined in Table 5-3.

Once the gases reach the nozzle, they experience an adiabatic, reversible
expansion process which is accompanied by a major drop in temperature and
pressure and a conversion of thermal energy into kinetic energy. Several increas-
ingly more complicated methods have been used for the analysis of the process.
For the simple case of frozen equilibrium and one-dimensional flow the state of
the gas throughout expansion in the nozzle is fixed by the entropy of the sys-
tem, which is presumed to be invariant as the pressure is reduced to the value
assigned to the nozzle exit plane. All the assumptions listed in Chapter 3 for an
ideal rocket would be valid here. Again, the effects of friction, divergence angle,
heat losses, shock waves, or nonequilibrium are neglected in the simple cases
but are considered in the more sophisticated solutions. The condensed (liquid or
solid) phases are similarly assumed to have zero volume and to be in kinetic
as well as thermal equilibrium with the gas flow. This implies that particles or
droplets are very small in size, move at the same velocity as the gas stream, and
have the same temperature as the gas at all places in the nozzle.

The chemical equilibrium during expansion in the nozzle can be analytically
regarded in the following ways:

1. When the composition is invariant throughout the nozzle, there are no
chemieal reactions or phase changes and the product composition at the
nozzle exit is identical to that of its chamber condition. The results are
known as frozen equilibrium rocket performance. This method usually is
simple, but underestimates the performanee, typically by 1 to 4%.

5.3. ANALYSIS OF NOZZLE EXPANSION PROCESSES

171

2. Instantaneous chemical equilibrium among all molecular species is main-
tained under the continuously variable pressure and temperature conditions
of the nozzle expansion process. Thus the product composition shifts; sim-
ilarly, instantaneous chemical reactions, phase changes or equilibria occur
between gaseous and condensed phases of all species in the exhaust gas.
The results so calculated are called shifting equilibrium performance. The
gas composition mass percentages are different in the chamber and the
nozzle exit. This method usually overstates the performance values, such
as c* or /j, typically by 1 to 4%. Here the analysis is more complex.

3. Chemical reactions do not occur instantaneously; even though the reac-
tions occur rapidly, they require a finite time. Reaction rates of specific
reactions can be estimated; the rates are usually a function of tempera-
ture, the magnitude of deviation from the equilibrium molar composition,
and the nature of the chemicals or reactions involved. The values of T,
c*, or Is for these types of equilibrium analysis usually are between those
of frozen and instantaneously shifting equilibria. This approach is almost
never used because of the lack of good data on reaction rates with multiple
simultaneous chemical reactions.

For an axisymmetric nozzle, both one-and two-dimensional analyses can be
used. The simplest nozzle flow analysis is one dimensional, which means that
all velocities and temperatures or pressures are equal at any normal cross section
of an axisymmetric nozzle. It is often satisfactory for preliminary estimates. In a
two-dimensional analysis the velocity, temperature, density, and/or Mach number
do not have a flat profile and vary somewhat over the cross sections. For nozzle
shapes that are not bodies of revolution (e.g., rectangular, scarfed, or elliptic) a
three-dimensional analysis should be performed.

If solid particles or liquid droplets are present in the nozzle flow and if the
particles are larger than about 0.1 pm average diameter, there will be a thermal
lag and velocity lag. The solid particles or liquid droplets do not expand like
a gas; their temperature decrease depends on losing energy by convection or
radiation, and their velocity depends on the drag forces exerted on the particle.
Larger-diameter droplets or particles are not accelerated as rapidly as the smaller
ones and flow at a velocity lower than that of the adjacent accelerating gas. Also,
the particulates are hotter than the gas and provide heat to the gas. While these
particles contribute to the momentum of the exhaust mass, they are not as efficient
as an all-gaseous exhaust flow. For composite solid propellants with aluminum
oxide particles in the exhaust gas, the loss due to particles could typically be 1 to
3%. The analysis of a two- or three-phase flow requires knowledge of an assump-
tion about the nongaseous matter, the sizes (diameters), size distribution, shape
(usually assumed to be spherical), optical surface properties (for determining the
emission/absorption or scattering of radiant energy), and their condensation or
freezing temperatures. Some of these parameters are not well known. Performance
estimates of flows with particles are explained in Section 3.5.

1 72 CHEMICAL ROCKET PROPELLANT PEREORMANCE ANALYSIS

TABLE 5-3. Typical Steps and Alternatives in the Analysis of Rocket Thermochemical
Processes in Nozzles

Step

Process Method/Implication/Assumption

Nozzle inlet
condition

Nozzle

expansion

Same as chamber exit; need to
know T\,pi,v\,H ,c* , p\, etc.

An adiabatic process, where flow
is accelerated and thermal
energy is converted into kinetic
energy. Temperature and
pressure drop drastically.

Several different analyses have
been used with different specific
effects. Can use one-, two-, or
three-dimensional flow pattern.

For simpler analyses assume the
flow to be uniformly mixed and
steady.

1 . Simplest method is inviscid isen-
tropic expansion flow with con-
stant entropy.

2. Include internal weak shock
waves; no longer a truly isentropic
process.

3. If solid particles are present, they
will create drag, thermal lag,
and a hotter exhaust gas. Must
assume an average particle size
and optical surface properties of
the particulates. Flow is no longer
isentropic.

4. Include viscous boundary layer
effects and/or nonuniform veloc-
ity profile.

Often a simple single correction factor is used with one-dimensional analyses to
correct the nozzle exit condition for items 2, 3, and/or 4 above. Computational fluid
dynamic codes with finite element analyses have been used with two- and
three-dimensional nozzle flow.

Chemical

equilibrium

during

nozzle

expansion

Due to rapid decrease in T and p, \
the equilibrium composition can
change from that in the
chamber. The four processes
listed in the next column allow
progressively more realistic
simulation and require more
sophisticated techniques.

Frozen equilibrium; no change in
gas composition; usually gives
low performance.

2. Shifting equilibrium or instan-
taneous change in composition;
usually overstates the perfor-
mance slightly.

3. Use reaction time rate analysis to
estimate the time to reach equi-
librium for each of the several
chemical reactions; some rate
constants are not well known;
analysis is more complex.

4. Use different equilibrium analysis
for boundary layer and main invis-
cid flow; will have nonuniform
gas temperature, composition, and
velocity profiles.

5.3. ANALYSIS OF NOZZLE EXPANSION PROCESSES 1 73

TABLE 5-3. (Continued)

Step Process Method/Implication/Assumption

Heat release in
nozzle

Nozzle shape
and size

Gas properties

Nozzle exit
conditions

Calculate

specific

impulse

Recombination of dissociated
molecules (e.g., H + H = H 2 )
and exothermic reactions due to
changes in equilibrium
composition cause an internal
heating of the expanding gases.
Particulates release heat to the
gas.

Can use straight cone, bell-shaped,
or other nozzle contour; bell
can give slightly lower losses.
Make correction for divergence
losses and nonuniformity of
velocity profile.

The relationships governing the
behavior of the gases apply to
both nozzle and chamber
conditions. As gases cool in
expansion, some species may
condense.

Will depend on the assumptions
made above for chemical
equilibrium, nozzle expansion,
and nozzle shape/contour.
Assume no jet separation.
Determine velocity profile and
fhe pressure profile af the
nozzle exit plane. If pressure is
not uniform across a section it
will have some cross flow.

Can be determined for different
altitudes, pressure ratios,
mixture ratios, nozzle area
ratios, etc.

Heat released in subsonic portion
of nozzle will increase the exit
velocity. Heating in the
supersonic flow portion of
nozzle can increase the exit
temperature but reduce the exit
Mach number.

Must know or assume a particular
nozzle configuration. Calculate
bell contour by method of
characteristics. Use Eq. 3-34
for divergence losses in conical
nozzle. Most analysis programs
are one- or two-dimensional.
Unsymmetrical nonround
nozzles may need
three-dimensional analysis.

Either use perfect gas laws or, if
some of the gas species come
close to being condensed, use
real gas properties.

Need to know the nozzle area
ratio or nozzle pressure ratio.
For quasi-one-dimensional and
uniform nozzle flow, see Eqs.
3-25 and 3-26. If V 2 is not
constant over the exit area,
determine effective average
values of V 2 and p 2 - Then
calculate profiles of T, p, efc.
For nonuniform velocify profile,
fhe solufion requires an iterative
approach. Can calculate the gas
conditions (T, p, etc.) at any
point in the nozzle.

Can be determined for average
values of V 2 ,P 2 , and p^ based
on Eqs. 2-6 or 2-14.

1 74 CHEMICAL ROCKET PROPELLANT PEREORMANCE ANALYSIS

The viscous boundary layer next to the nozzle wall has velocities substantially
lower than that of the inviscid free stream. The slowing down of the gas flow near
the wall due to the viscous drag actually causes the conversion of kinetic energy
back into thermal energy, and thus some parts of the boundary layer can be hotter
than the local free-stream static temperature. A diagram of a two-dimensional
boundary layer is shown in Figure 3-16. With turbulence this boundary layer
can be relatively thick in large-diameter nozzles. The boundary layer is also
dependent on the axial pressure gradient in the nozzle, the nozzle geometry,
particularly in the throat region, the surface roughness, or the heat losses to the
nozzle walls. Theoretical boundary layer analyses with unsteady flow are only
approximations, but are expected to improve in the future as our understanding of
the phenomena and computational fluid dynamics (CFD) techniques are validated.
The net effect is a nonuniform velocity and temperature profile, an irreversible
friction process in the viscous layers, and therefore an increase in entropy and a
slight reduction (usually less than 5%) of the kinetic exhaust energy. The layers
immediately adjacent to the nozzle walls have laminar and subsonic flow.

At the high combustion temperatures a small portion of the combustion gas
molecules dissociate (split into simpler species); in this dissociation process some
energy is absorbed. When energy is released during reassociation (at lower pres-
sures and temperatures in the nozzle), this reduces the kinetic energy of the
exhaust gas at the nozzle exit. This is discussed further next.

For propellants that yield only gaseous products, extra energy is released in
the nozzle, primarily from the recombination of free-radical and atomic species,
which become unstable as the temperature is decreased in the nozzle expansion
process. Some propellant products include species that condense as the tem-
perature drops in the nozzle expansion. If the heat release on condensation is
large, the difference between frozen and shifting equilibrium performance can be
substantial.

In the simplest method the exit temperature T 2 is determined for an isen-
tropic process (frozen equilibrium) by considering the entropy to be constant.
The entropy at the exit is the same as the entropy in the chamber. This deter-
mines the temperature at the exit and thus the gas condition at the exit. From
the corresponding change in enthalpy it is then possible to obtain the exhaust
velocity and the specific impulse. For those analysis methods where the nozzle
flow is not really isentropic and the expansion process is only partly reversible,
it is necessary to include the losses due to friction, shock waves, turbulence, and
so on. The result is a somewhat higher average nozzle exit temperature and a
slight loss in /^ . A possible set of steps used for the analysis of nozzle processes
is given in Table 5-3.

When the contraction between the combustion chamber (or the port area) and
the throat area is small (Ap/A, < 3), the acceleration of the gases in the chamber
causes a drop in the effective chamber pressure at the nozzle entrance. This
pressure loss in the chamber causes a slight reduction of the values of c and 4 .
The analysis of this chamber configuration is treated in Ref. 5-14 and some data
are briefly shown in Table 3-2.

5.4. COMPUTER-ASSISTED ANALYSIS 175

5.4. COMPUTER-ASSISTED ANALYSIS

All the analyses discussed in this chapter are done today with computer programs.
Most are based on minimizing the free energy. This is a simpler approach than
relying on equilibrium constants, which was used some years ago. Once the
values of nj and Ti are determined, it is possible to calculate the molecular mass
of the gases (Eq. 5-5), the average molar specific heats Cp by a similar formula,
and the specific heat ratio k from Eqs. 3-6 and 5-7. This then characterizes
the thermodynamic conditions in the combustion chamber. With these data we
can calculate c*, R, and other mixture parameters of the chamber combustion.
The nozzle expansion process simulated with computers give the performance
(such as Is, c, or A 2 /A,) and the gas conditions in the nozzle; it usually includes
several of the corrections mentioned in Chapter 3. Programs exist for one-, two-,
and three-dimensional flow patterns.

More sophisticated solutions include a supplementary analysis of combustion
chamber conditions where the chamber velocities are high (see Ref. 5-14), a
boundary layer analysis, a heat transfer analysis, or a two-dimensional axisym-
metric flow with nonuniform flow properties across any cross section of the
nozzle. Time-dependent chemical reactions in the chamber are usually neglected,
but they can be analyzed by estimating the time rate at which the reaction occurs.
This is described in Ref. 5-3.

A commonly used computer program, based on equilibrium compositions,
has been developed at the NASA Glenn Laboratories and is known as the
NASA CEA code (Chemical Equilibrium with Applications). It is described in
Ref. 5-13, Vols. 1 and 2. Key assumptions for this program are one-dimensional
forms of the continuity, energy and momentum equations, negligible velocity at
the forward end of the combustion chamber, isentropic expansion in the noz-
zle, ideal gas behavior, and chemical equilibrium in the combustion chamber. It
includes options for frozen flow and for narrow chambers (for liquid propellant
combustion) or port areas with small cross sections (for solid propellant grains),
where the chamber flow velocities are relatively high, resulting in extra pressure
losses and a slight loss in performance. NASA’s CEA code has recently become
part of a commercial code named Cequel™, which extends the code’s original
capabilities.

Other relatively common computer codes used in the United States for ana-
lyzing the converging-diverging nozzle flow include:

ODE (one-dimensional equilibrium code) which features infinitely fast chem-
ical reactions (shifting equilibrium) and includes all gaseous constituents.

ODK (one-dimensional kinetics) which incorporates finite chemical reaction
kinetic rates for temperature-dependent composition changes in the flow
direction with uniform flow properties across any nozzle section. It is used
as a module in more complex codes and has no provision for particles.

TDK (two-dimensional kinetic code) which incorporates finite kinetic chemical
reaction rates and radial variance in flow properties. It has no provision for
particles.

1 76 CHEMICAL ROCKET PROPELLANT PEREORMANCE ANALYSIS

VIPERP ( viscous interaction performance evaluation routine for two-phased
flows) a parabolized Navier-Stokes code for internal two-phased nozzle
flow with turbulent and nonequilibrium reacting gases. It can be used with
solid particles but requires data (or assumptions) on the amount of solids,
particle size distribution, or their shape (see, e.g., pp. 503 to 505 in 7th
Edition).

More information on these computer codes may be obtained from the appro-
priate government offices and/or from small companies (who actually run the
nozzle codes for their customers). Some of the more sophisticated codes are
proprietary to propulsion organizations and not publicly available.

5.5. RESULTS OF THERMOCHEMICAL CALCULATIONS

Extensive computer results of these machine calculations are available and only a
few samples are indicated here to illustrate typical effects of the variations of key
parameters. In general, high specific impulse or high values of c* can be obtained
if the average molecular mass of the reaction products is low (usually this implies
a formulation rich in hydrogen) and/or if the available chemical energy (heat of
reaction) is large, which means high combustion temperatures (see Eq. 3-16).

Table 5-4 shows calculated results for a liquid oxygen, liquid hydrogen thrust
chamber taken from an example of this reference. It has shifting equilibrium in
the nozzle flow. The narrow chamber has a cross section that is only a little
larger than the throat area. The large pressure drop in the chamber (approximately
126 psi) is due to the energy needed to accelerate the gas, as discussed in Section
3.3 and Table 3-2.

Values of calculated specific impulse will be higher than those obtained from
firing actual propellants in rocket units. In practice it has been found that the
experimental values are about 3 to 12% lower than those calculated by the
method explained in this chapter. Because the nozzle inefficiencies explained in
Chapter 3 must be considered, only a portion of this correction (perhaps 1 to
4%) is due to combustion inefficiencies.

Eigures 5-1 to 5-6 indicate the results of performance calculations for the liq-
uid propellant combination, liquid oxygen-RP-1. These data are taken from Refs.
5-7 and 5-8. The RP-1 fuel is a narrow-cut hydrocarbon similar to kerosene
with an average of 1.953 mol of hydrogen for each mole of carbon; thus it has
a nominal formula of CHi 953 . The calculations are limited to a chamber pres-
sure of 1000 psia. Most of the curves are for optimum area ratio expansion to
atmospheric pressure, namely 1 atm or 14.696 psia, and for a limited range of
oxidizer-to-fuel mixture ratios.

Eor maximum specific impulse. Figs. 5-1 and 5-4 show an optimum mixture
ratio of approximately 2.3 (kg/sec of oxidizer flow divided by kg/sec of fuel
flow) for frozen equilibrium expansion and 2.5 for shifting equilibrium with
gas expansion to sea-level pressure. The maximum values of c* are at slightly

0^ On

O -H —

r- ra

o r- ON
lO 00 o

'Tf <T)

OOCNOOOOOO

OOiO'TfOOOO

OOOOnOOOO

OOrONOOOOO

doddoodo

!/3

(N O
<N <N ra 00 o

(N 00 (T>
00 NO —
«N

(NOOOOOOOO

OOiO'^fOOOO

OOOONOOOO

oomNOOOOO

dddddddd

(N (N
NO d

r- ON

(N IT) 00 (N
‘ CN lO

'^or-iooo'^o

(Nommoooo

OOOONOOOO

OOrONOOOOO

dddddddd

D NO r--

15 r-- 00 o «N

X O — ON — H (N

W doNco — dddincN

On cm
lO lO
NO ON
On 00
(N NO

o ON r- ON
o o r- o
0 0^0
o o o o
o o o o

oooooooo

biD

•c

o o o
lo CO o o
00 — lo O O

On r--
r- <T)
00 IT)

— 00 O —

C-; — <N p
On <N — 00 d X

00 o o
lO m ‘

ro m
On IT)

(N X

O O (N O
O O O O

S300000000

00
00 lo
m 00
On m
(N X

O (N O -H
O O ro O

o o o o
dddddddd

o rs o CO

On O 'Tf

ro O X
ro O On ro
O O (N X

O O ro O
o o o o

oooooooo

'2 ^
C3 ^
<U

• ^ ^
^ _g

d §

^ B

X c

'X, 8

s; t-H

(U c/5

C«

^ -j; o 3

O «
S ffi

' s

K N
O O

177

‘'Chamber contraction ratio Ai/A,.

*If cut off at throat.

c is the effective exhaust velocity in a vacuum.

V 2 is the nozzle exit velocity at optimum nozzle expansion.
NA means not applicable.

178

CHEMICAL ROCKET PROPELLANT PEREORMANCE ANALYSIS

1850-
S 1800-

I/)

o 1750-
1700-

Mixture ratio (oxidizer/fuel)

FIGURE 5-1. Calculated performance analysis of liquid oxygen and hydrocarbon fuel
as a function of mixture ratio.

different mixture ratios. This optimum mixture ratio is not the value for highest
temperature, which is usually fairly close to the stoichiometric value. The stoi-
chiometric mixture ratio is more than 3.0 where much of the carbon is burned to
CO 2 and almost all of the hydrogen to H 2 O.

Because shifting equilibrium makes more enthalpy available for conversion to
kinetic energy, it gives higher values of performance (higher Is or c*) and higher
values of nozzle exit temperature for the same exit pressure (see Fig. 5-1). The
influence of mixture ratio on chamber gas composition is evident from Fig. 5-2.
A comparison with Fig. 5-3 indicates the marked changes in the gas composition
as the gases are expanded under shifting equilibrium conditions. The influence of
the degree of expansion, or of the nozzle exit pressure on the gas composition,
is shown in Fig. 5-6. As the gases are expanded to higher area ratios and lower
exit pressure (or higher pressure ratios) the performance increases; however.

5.5. RESULTS OF THERMOCHEMICAL CALCULATIONS 179

FIGURE 5-2. Calculated chamber gas composition for liquid oxygen and hydrocarbon
fuel as a function of mixture ratio. Aggressive gases, such as O 2 , O, or OH, can cause
oxidation of the wall materials In the chamber and the nozzle.

the relative increase diminishes as the pressure ratio is further increased (see
Figs. 5-5 and 5-6).

Dissociation of molecules absorbs considerable energy and causes a decrease
in the combustion temperature, which in turn can reduce the specific impulse. Dis-
sociation of the reaction products increases as the chamber temperature rises, and
decreases with increasing chamber pressure. Atoms or radicals such as monatomic
O or H and OH are formed, as can be seen from Fig. 5-2; some unreacted O 2 also
remains at the higher mixture ratios and very high combustion temperatures. As
the gases are cooled in the nozzle expansion, the dissociated species react again
to form molecules and release heat into the flowing gases. As can be seen from
Fig. 5-3, only a small percentage of dissociated species persists at the nozzle
exit and only at the high mixture ratio, where the exit temperature is relatively
high. (See Fig. 5-1 for exit temperatures with shifting equilibria). Heat released
in a supersonic flow actually reduces the Mach number.

180

CHEMICAL ROCKET PROPELLANT PEREORMANCE ANALYSIS

Mixture ratio (oxidizer/fuelt

FIGURE 5-3. Calculated nozzle exit gas composition for shifting equilibrium conditions
as a function of mixture ratio. Breakdown into O, OH, or H and free O 2 occurs only at
the higher temperatures or higher mixture ratios.

Results of calculations for several different liquid and solid propellant combi-
nations are given in Tables 5-5 and 5-6. For the liquid propellant combinations,
the listed mixture ratios are optimum and their performance is a maximum. For
solid propellants, practical considerations (such as propellant physical properties)
do not always permit the development of a satisfactory propellant grain when the
ingredients are mixed in optimum performance proportions (insufficient binder);
therefore the values listed for solid propellants in Table 5-6 correspond in part
to practical formulations with reasonable physical and ballistic properties.

Calculated data obtained from Ref. 5-13 are presented in Tables 5-7 to 5-9
for a specific solid propellant to indicate typical variations in performance or
gas composition. This particular propellant consists of 60% ammonium perchlo-
rate (NH 4 CIO 4 ), 20% pure aluminum powder, and 20% of an organic polymer
of an assumed chemical composition, namely C 3 . 1 ON 0 . 84 FI 5 Table 5-7 shows

5.5. RESULTS OF THERMOCHEMICAL CALCULATIONS 181

1.8 2.0 2.2 2.4 2.6 2.8 3.0

Mixture ratio (oxidizer/fuel)

FIGURE 5-4. Variation of theoretical specific impulse with mixture ratio and pressure
ratio, calculated for frozen equilibrium.

the variation of several performance parameters with different chamber pressures
expanding to atmospheric exit pressure. The area ratios listed are optimum for
this expansion with shifting equilibrium. The exit enthalpy, exit entropy, thrust
coefficient, and the specific impulse also consider shifting equilibrium conditions.
The characteristic velocity c* and the chamber molecular mass are functions of
chamber conditions only. Table 5-8 shows the variation of gas composition with
chamber pressure. Some of the reaction products are in the liquid phase, such
as AI 2 O 3 . Table 5-9 shows the variation of nozzle exit characteristics and com-
position for shifting equilibria as a function of exit pressure or pressure ratio
for a fixed value of chamber pressure. Table 5-9 shows how the composition is
shifted during expansion in the nozzle and how several of the species present in
the chamber have disappeared at the nozzle exit. These three tables show theo-
retical results calculated on a computer; some of the thermodynamic properties
of the reactants and reaction products probably do not warrant the indicated high
accuracy of five significant figures. In the analysis for chemical ingredients of this
solid propellant, approximately 76 additional reaction products were considered

002 ^ I— 1 — I — I 1 i 1 M 1 1 1 — I Mill I I I 1 I 1 I I I I ozfr

)U3pi|^ao3 )snjqx

^ 00 CM vp o

CM ^ ^ o o

I — ^ ^ 1

>1. ‘ajnjejadujax

D8S ’8S|ndLU! ojjpeds

I I I I I rT T I I 1 — I — I 1 1

8 8SS§ S oooio^ cv,

0|)ej eajy

182

FIGURE 5-5. Variation of calculated parameters with pressure ratio for liquid oxygen-hydrocarbon propellant at a mixture ratio of 2 . 20 . An
increase in pressure ratio is due to an increase in chamber pressure, a decrease of nozzle exit pressure (larger area ratio and higher altitude), or
both.

)ua3J3d 3|oai ‘uoDisoduioQ

183

FIGURE 5-6. Variation of exhaust gas composition at nozzle exit with pressure ratio at a fixed mixture ratio and for shifting equilibrium. For
frozen equilibrium the composition would be the same as in the chamber, as shown in Fig. 5-2.

TABLE 5-5. Theoretical Performance of Liquid Rocket Propellant Combinations

B •

X *

B D.

X S

2 '5

O 0)

<C 00

CQ —

>»

O lO
O Oh
ro (N

lo Oh
X 00

m m

00 00 CN

r- lo r-

fNl (N (N

Oh IT)

mroOhOOhrooomiO'^OhooioOhOhO'

lom — CN00(N’^OiO'^00 00'^0hin(N(N —
roiot^OhCNmr^om\0(NOm'^hOooir) —
oooooooo'Tf'^r^oooooo-HiNioior^r^hor^

— — -h(N(N(N(N — -H — —

I fo hO oo hO

^ Oh fs -H

Oh Oh O 00 o

r- in in ho ho r-

hOhOin’^OhOh-Hr-'(N'^rofnoooo(NCN<NOhooinor--(NO
cscMoooinOhr^r^'^Ohin — oooinin"^r^OhOhOr^mOhr^ho
inin(N’^OhOhinhOinininc^oOhfNi-H(Ncn(Nfn(N-H(NOh-Hr^

mmfnm(N(Nmrofnm'^'^mrocnmmcnromcnmmmm<N

^Ohor^hooo^(NhOooOh — minoraoo — mohoincomr^oh
ooooop<N(NppOhOh(Nfnrn'^fNl<N— -CNlcNCN^rncncacNC^
oo — ^c30^-^oo— 'c30^-^—

' O O O O '

oO'^oo<N’^hC)OhinroO'^ooo'^(NO inin mo
(Npc^p'^pcNinmppminppmpp'^t— -p-Hoqr-(NO
mmoom-^c^tN — -H—

Urn

•o

a-s

X N

S S

Q?,

Os:

^ ^ 2 S

-I

^ ^ o'

in in 05

S o

V. 'c

00^

>. '

184

Notes:

Combustion chamber pressure — 1000 psia (6895 kN/m^); nozzle exit pressure — 14.7 psia (1 atm); optimum expansion.
Adiabatic combustion and isentropic expansion of ideal gas.

The specific gravity at the boiling point was used for those oxidizers or fuels that boil below 20°C at 1 atm pressure.
Mixture ratios are for approximate maximum value of Is .

5.5. RESULTS OF THERMOCHEMICAL CALCULATIONS 185

TABLE 5-6. Theoretical Performance of Typical Solid Rocket Propellant Combinations

Oxidizer

Fuel

(g/cm^)“

(K)

(m/sec)*

(kg/mol)

(sec)*

Ammonium

nitrate

1 1 % binder and
7% additives

1.51

1282

1209

20.1

192

1.26

Ammonium

perchlorate

78-66%

18% organic
polymer binder
and 4-20%
aluminum

1.69

2816

1590

25.0

262

1.21

Ammonium

perchlorate

84-68%

12% polymer
binder and
4-20%
aluminum

1.74

3371

1577

29.3

266

1.17

“Average specific gravity of solid propellant.

^Conditions for f and c* : Combustion chamber pressure: 1000 psia; nozzle exit pressure: 14.7
psia; optimum nozzle expansion ratio; frozen equilibrium.

in addition to the major product species. This includes, for example, CN, CH,
CCl, Cl, NO, and so on. Their calculated mole fractions were very small and
therefore they have been neglected and are not included in Table 5-8 or 5-9.

Calculations of this type are useful in estimating performance {Is,c*,Cf,e,
etc.) for a particular chamber pressure and nozzle exit pressure, and knowl-
edge of the gas composition, as indicated by the previous figures and tables,
permits a more detailed estimate of other design parameters, such as gas-hlm
properties for heat transfer determination, radiation characteristics of the flame
inside and outside the thrust chambers, and the acoustic characteristics of the
gases. Performance data relevant to hybrid propellants are presented briefly in
Chapter 16.

The thermochemical analysis of this chapter can also be applied to gas gener-
ators; the results (such as gas temperature Ti, the specific heat Cp, specific heat
ratio k, or composition) are used for estimating turbine inlet conditions or tur-
bine power. In gas generators and prebumers for staged combustion cycle rocket
engines (explained in Section 6.6) the gas temperatures are much lower, to avoid
damage to the turbine blades. Typically, the combustion reaction gases are at 800
to 1200 K, which is lower than the gas in the thrust chamber (2900 to 3600 K).
Examples are listed in Table 5-10 for a chamber pressure of 1000 psia. Some
species in the gases will not be present (such as atomic oxygen or hydroxyl),
and often real gas properties will need to be used because some of these gases
do not behave as a perfect gas at these temperatures.

Q-

< O'

c ®

s:
fc ^

a> 43

c3 ^

Oh a

U td

§ -

1 "p

£ s

u ->

^ g
S 2

’H c

‘cfl

u g

f I 2

o ^
® X
C PJ

.9 Vh

5 (u
S

r- p

I £

M (U

H U

»o

1— 1

»o

GTn

CM*

»o

1-H

»o

i> r-

CM*

»n

1—

-H

CM*

1-H

<—

»n

'sO o^

CM*

»n

2 c

III

D a (D

Cfl Q.
V3 Cfl ^

D (U 5
^ ^

G< Ph ii

u Vh ;-H
D (U (U
43 43 43

see

cd cd

43 4= -C

u u u

D cd

s ^

D >,
Oh a
E cd

oi)

0^3
^ Po o
^ 3 ®

43 ^

^ (U

i S 's

.32 i3 cd
X C 43

PJ PP u

h ^

p !Z3

B ^

I =

y 3
1 >

p o
o ^
S S

'S '?<

w w

C ^
C4 -i?

X P
s

5 'S

D c.
D- -.3

S E

o o
’o 'o

dJ d)

a Oh
CO CO

D 2
Dh ^

CO U

186

Source: From Ref. 5-13.

5.5. RESULTS OF THERMOCHEMICAL CALCULATIONS 187

TABLE 5-8. Mole Fraction Variation of Chamber Gas Composition with Combustion
Chamber Pressure for a Solid Propellant

Pressure (psia)

1500

1000

750

500

200

Pressure (atm) or

102.07

68.046

51.034

34.023

13.609

pressure ratio
Ingredient
A1

0.00007

0.00009

0.00010

0.00012

0.00018

AlCl

0.00454

0.00499

0.00530

0.00572

0.00655

AlClz

0.00181

0.00167

0.00157

0.00142

0.00112

AICI3

0.00029

0.00023

0.00019

0.00015

0.00009

AlH

0.00002

AlO

0.00007

0.00009

0.00011

0.00013

0.00019

AlOCl

0.00086

0.00095

0.00102

0.00112

0.00132

AlOH

0.00029

0.00032

0.00034

0.00036

0.00041

AIO2H

0.00024

0.00026

0.00028

0.00031

0.00036

AI2O

0.00003

0.00004

0.00005

0.00006

AI2O3 (solid)

0.00000

AI2O3 (liquid)

0.09425

0.09378

0.09343

0.09293

0.09178

0.22434

0.22374

0.22328

0.22259

0.22085

COCl

0.00001

0.00000

CO2

0.00785

0.00790

0.00793

0.00799

0.00810

0.00541

0.00620

0.00681

0.00772

0.01002

CI2

0.00001

0.02197

0.02525

0.02776

0.03157

0.04125

HCl

0.12021

0.11900

0.11808

0.11668

0.11321

HCN

0.00003

0.00002

0.00001

0.00000

HCO

0.00003

0.00002

0.00001

0.32599

0.32380

0.32215

0.31968

0.31362

H2O

0.08960

0.08937

0.08916

0.08886

0.08787

NH2

0.00001

0.00000

NH3

0.00004

0.00003

0.00002

0.00001

0.00019

0.00021

0.00023

0.00025

0.00030

0.09910

0.09886

0.09867

0.09839

0.09767

0.00010

0.00014

0.00016

0.00021

0.00036

0.00262

0.00297

0.00324

0.00364

0.00458

0.00001

0.00002

0.00004

Source: From Ref. 5-13.

TABLE 5-9. Calculated Variation of Thermodynamic Properties and Exit Gas Composition for an Aluminized Perchlorate Propellant with
p] = 1500 psia and Various Exit Pressures at Shifting Equilibrium and Optimum Expansion

1-H

»o

— H

»o

— H

<—

CNJ

lO)

1 -H

CnJ

» 0 )

»o

CLh Oh

(U 3

is -2

0 D

N e-

N C

1 I

1 -iH

!Z3

C/3 3

0 -

C'l

Gh ’0

188

AI 2 O 3 (solid) 0.00000 0.00000 0.09955 0.09969 0.09974 0.09976 0.09976

AI 2 O 3 (liquid) 0.09425 0.09608 0.00000 0.00000 0.00000 0.00000 0.00000

CO 0.22434 0.22511 0.22553 0.22416 0.22008 0.21824 0.21671

CO 2 0.00785 0.00787 0.00994 0.01126 0.01220 0.01548 0.01885

Cl 0.00541 0.00441 0.00074 0.00028 0.00009 0.00002 0.00000

H 0.02197 0.01722 0.00258 0.00095 0.00030 0.00007 0.00001

HCl 0.12021 0.12505 0.13635 0.13707 0.13734 0.13743 0.13746

CNl

o in

O 'O

o o
o o

O -H

d d

CNJ

— H

r--

1-^

CNJ

— H

n ^ O C4 K
K Z Z O O

189

190 CHEMICAL ROCKET PROPELLANT PEREORMANCE ANALYSIS

TABLE 5-10. Typical Gas Characteristics for Fuel-Rich Liquid Propellant Gas
Generators

Propellant

Ti (K)

Gas Constant R
(ft-lbf/lbm-°r)

Oxidizer-to-
Fuel Ratio

Specific Heat
Cp (kcal/kg-K)

Liquid oxygen

900

1.370

421

0.919

1.99

and liquid
hydrogen

1050

1.357

375

1.065

1.85

1200

1.338

347

1.208

1.78

Liquid oxygen

900

1.101

45.5

0.322

0.639

and kerosene

1050

1.127

55.3

0.423

0.654

1200

1.148

64.0

0.516

0.662

Nitrogen

1050

1.420

87.8

0.126

0.386

tetroxide and

dimethyl

hydrazine

1200

1.420

99.9

0.274

0.434

PROBLEMS

1. Explain the physical or chemical reasons for a maximum value of specific impulse at
a particular mixture ratio of oxidizer to fuel.

2. Explain why, in Table 5-8, the relative proportion of monatomic hydrogen and mona-
tonic oxygen changes markedly with different chamber pressures and exit pressures.

3. This chapter contains several charts for the performance of liquid oxygen and RP-1
hydrocarbon fuel. By mistake the next shipment of cryogenic oxidizer contains at least
15% liquid nitrogen. Explain what general trends should be expected in the results
of the next test in the performance values, the likely composition of the exhaust gas
under chamber and nozzle conditions, and the optimum mixture ratio.

4. A mixture of perfect gases consists of 3 kg of carbon monoxide and 1.5 kg of nitrogen
at a pressure of 0.1 MPa and a temperature of 298.15 K. Using Table 5-1, find (a) the
effective molecular mass of the mixture, (b) its gas constant, (c) specific heat ratio,
(d) partial pressures, and (e) density.

Answers: (a) 28 kg/kg-mol, (b) 297 J/kg-K, (c) 1.40, (d) 0.0666 and 0.0333 MPa, (e)
1.13 kg/m^

5. Using information from Table 5-2, plot the value of the specific heat ratio for carbon
monoxide (CO) as a function of temperature. Notice the trend of this curve; it is typical
of the temperature behavior of other diatomic gases.

Answers: k = 1.22, at 3500 K, 1.30 at 2000 K, 1.39 at 500 K.

6. Modify and tabulate two entries in Table 5-5 for operation in the vacuum of space,
namely oxygen/hydrogen and nitrogen tetroxide/hydrazine. Assume the data in the
table represents the design condition.

SYMBOLS 191

7. Various experiments have been conducted with a liquid monopropellant called
nitromethane (CH 3 NO 2 ), which can be decomposed into gaseous reaction products.
Determine the values of T, 971, k, c*,Cp, and /j using the water-gas equilibrium
conditions. Assume no dissociations and no O 2 .

Answers: 2470 K, 20.3 kg/kg-mol, 1.25, 2394 m/sec, 1.57, 244 sec.

8. The figures in this chapter show several parameters and gas compositions of liquid
oxygen burning with RP-1, which is a kerosene-type material. For a mixture ratio of
2.0, use the compositions to verify the molecular mass in the chamber and the specific
impulse (frozen equilibrium flow in nozzle) in Fig. 5-1.

SYMBOLS

(Symbols referring to chemical elements, compounds, or mathematical operators
are not included in this list.)

a number of kilogram atoms

A, throat area, m^

c* characteristic velocity, m/sec

Cp specific heat per unit mass, J/kg-K

Cp molar specific heat at constant pressure of gas mixture, J/kg-mol-K

go acceleration of gravity at sea level, 9.8066 m/sec^

G Gibbs free energy for a propellant combustion gas mixture, J/kg

AfG^ change in free energy of formation at 298.15 K and 1 bar
Gj free energy for a particular species j, J/kg

AH overall enthalpy change, J/kg or J/kg-mol

AHj enthalpy change for a particular species j , J/kg
ArH^ heat of reaction at reference 298.15 K and 1 bar, J/kg
AfH^ heat of formation at reference 298.15 K and 1 bar, J/kg
h enthalpy for a particular species, J/kg or J/kg-mol

Is specific impulse, sec

k specific heat ratio

m number of gaseous species

m mass flow rate, kg/sec

9JI molecular mass (also called molecular weight) of gas mixture, kg/mol

n total number of moles per unit mass (kg-mol/kg) of mixture

tij moles of species j, kg-mol/kg

P pressure of gas mixture, N/m^

R gas constant, J/kg-K

R' universal gas constant, 8314.3 J/kg mol-K

S entropy, J/kg mol-K

T absolute temperature, K

Tad adiabatic temperature, K

U internal energy, J/kg-mol

192 CHEMICAL ROCKET PROPELLANT PEREORMANCE ANALYSIS

V gas velocity, m/sec

V specific volume, m^/kg

Xj mole fraction of species j

Greek Letters

p density, kg/m^

Subscripts

a, b
c, d
i

mix

ref

molar fractions of reactant species A or B
molar fractions of product species C or D
atomic species in a specific propellant
constituents or species in reactants or products
mixture of gases

at reference condition (also superscript 0)
chamber condition
nozzle exit condition
ambient atmospheric condition

REFERENCES

5-1. F. Van Zeggeren and S. H. Storey, The Computation of Chemical Equilibria, Cam-
bridge University Press, Cambridge, 1970.

5-2. S. S. Penner, Thermodynamics for Scientists and Engineers , Addison- Wesley Pub-
lishing, Reading, MA, 1968.

5-3. S. I. Sandler, Chemical and Engineering Thermodynamics , John Wiley & Sons,
New York, 1999.

5-4. R. H. Dittman and M. W. Zemansky Heat and Thermodynamics , 7th ed., McGraw-
Hill Book Company, New York, 1996.

5-5. K. Denbigh, The Principles of Chemical Equilibrium, 4th ed., Cambridge Univer-
sity Press, Cambridge, 1981.

5-6. K. K. Kuo, Principles of Combustion, 2nd ed., John Wiley & Sons, Hoboken, NJ,
2005.

5-7. JANAE Thermochemical Tables, Dow Chemical Company, Midland, Ml, Series A
(June 1963) through Series E (January 1967).

5-8. M. W. Chase, C. A. Davies, J. R. Downey, D. J. Frurip, R. A. McDonald, and A.
N. Syverud, JANAE Thermochemical Tables, 3rd ed.. Part I, Journal of Physical
and Chemical Reference Data, Vol. 14, Supplement 1, American Chemical Society,
American Institute of Physics, and National Bureau of Standards, 1985.

REFERENCES 193

5-9. D. D. Wagman et al., “The NBS Tables of Chemical Thermodynamic Properties,”
Journal of Physical and Chemical Reference Data, Vol. 11, Supplement 2, Amer-
ican Chemical Society, American Institute of Physics, and National Bureau of
Standards, 1982.

5-10. J. B. Pedley, R. D. Naylor, and S. P. Kirby, Thermochemical Data of Organic
Compounds, 2nd ed.. Chapman & Hall, London, 1986.

5-11. B. J. McBride, S. Gordon, andM. Reno, “Thermodynamic Data for Fifty Reference
Elements,” NASA Technical Paper 3287, January 1993.

5-12. B. J. McBride and S. Gordon, “Computer Program for Calculating and Fitting
Thermodynamic Functions,” NASA Reference Publication 1271 , November 1992.

5-13. S. Gordon and B. J. McBride, “Computer Program for Calculation of Complex
Chemical Equilibrium Compositions and Applications, Vol. 1: Analysis” (October
1994) and “Vol. 2: User Manual and Program Description” (June 1996), NASA
Reference Publication 1311 .

5-14. S. Gordon and B. J. McBride, “Finite Area Combustor Theoretical Rocket Perfor-
mance,” NASA TM 100785, April 1988.

CHAPTER 6

LIQUID PROPELLANT ROCKET
ENGINE FUNDAMENTALS

This chapter gives an overview of liquid propellant rocket engines. It is the first
of six chapters devoted to this subject. It identifies the types of liquid rocket
engines, their key components, different propellants, and tank conhgurations.
It also discusses two types of feed systems, engine cycles, propellant tanks,
their pressurization subsystems, engine controls, valves, piping, and structure.
Chapter 7 covers liquid propellants in more detail. Chapter 8 describes thrust
chambers (and their nozzles), small thrusters, and heat transfer. Chapter 9 is
about the combustion process and Chapter 10 discusses turbopumps. Chapter 11
discusses engine design, engine controls, propellant budgets, engine balance and
calibration, and overall engine systems.

In this book a liquid propellant rocket propulsion system consist of a rocket
engine and a set of tanks for storing and supplying propellants. It has all the
hardware components and the propellants necessary for its operation, that is, for
producing thrust. See Refs. 6-1 and 6-2. They are described in this chapter.
A rocket engine consists of one or more thrust chambers, a feed mechanism
for feeding the propellants from their tanks into the thrust chamber(s), a power
source to furnish the energy for the feed mechanism, suitable plumbing or piping
to transfer the liquid propellants, a structure to transmit the thrust force, and
control devices (including valves) to start and stop and sometimes also vary the
propellant flow and thus the thrust. Propellants are either pushed out of their
tanks by high-pressure gas or they are pumped by pumps to the thrust chambers.
Figure 6-1 shows the Space Shuttle main engine, a large sophisticated high-
performance liquid propellant rocket engine. The index lists the pages where
information on the Space Shuttle, its main rocket engine, and various components
can be found.

194

195

FIGURE 6-1. Two views of the Space Shuttle Main Engine (SSME). Its flowsheet is in Figure 6-12 and some component data are in
Chapter 10. (Courtesy of Pratt & Whitney Rocketdyne and NASA.)

196 LIQUID PROPELLANT ROCKET ENGINE FUNDAMENTALS

In some applications rocket engines may also include a thrust vector con-
trol system (explained in Chapter 18), a random variable thrust feature (see
Section 8.5), an engine condition monitoring or engine health monitoring sub-
system (see Section 11.4), and various instrumentation/measuring devices (see
Chapter 21). The tanks, which store the propellants, and the subsystem for pres-
surizing the tanks with gas are in this chapter and are considered in this book to
be part of the rocket propulsion system.*

The design of any propulsion system is tailored to fit a specific mission require-
ment as explained in Chapter 19. These requirements are usually stated in terms
of the application, such as anti-aircraft missile or second stage of a space launch
vehicle, flight velocity increment, flight path and flight maneuvers, launch sites,
minimum life (in storage or in orbit), or number of vehicles to be delivered. The
requirement usually include constraints of the inert engine weight, cost, or safety
provisions. Other criteria, constraints and the selection process are in Chapter 19.

From the mission requirements and definition one can derive the propulsion
system and the engine requirements , which include the thrust-time prohle, the
minimum specific impulse, number of thrust chambers, total impulse, number
of restarts, if any, likely propellants, or constraints of engine masses or engine
sizes. Some engine parameters, such as thrust, chamber pressure, mixture ratio, or
nozzle exit area ratio, can be analytically optimized for a specific mission. Other
engine parameters can be selected based on experience and/or design studies,
including the feed system, the arrangement of the engine components, engine
cycle, thrust modulation, or alternate methods of thrust vector control. Two or
more preliminary or conceptual designs can be compared with the objective of
arriving at a selected propulsion design for the mission.

Tables 1-3, 11-2, and 11-3 give typical data for selected rocket engines.
Many different types of rocket engines have been studied, built, and flown, rang-
ing in thrust size from less than 0.01 Ibf to over 1.75 million pounds, with
one-time operation or multiple starts (some have over 150,000 restarts), with or
without thrust modulation (called throttling), single use or reusable, arranged as
single engines, or in clusters of multiple units.

One way to categorize liquid propellant rocket engines is described in
Table 6-1. There are two categories, namely those used for boosting a
payload and imparting a significant velocity increase to a payload and auxiliary
propulsion for trajectory adjustments and attitude control. Liquid propellant

*The responsibilities for the design, development, fabrication, and operation of a propulsion system
are usually shared between a rocket engine organization and a flight vehicle organization. However,
the allocations of the responsibilities for components or subsystems have not been rigid or consistent
in the literature and in actual industry practice. For example, some vehicle design/development
organizations have considered the tanks and parts of the engine structure to be really a part of their
vehicle. The tank pressurization system has in various scenarios been considered to be part of either
the engine, the propulsion system, or the vehicle. For some reaction control systems the vehicle
developer often assumed the responsibility for the propulsion systems and obtained only the small
thrusters and their small propellant valves from a rocket engine company. In some programs, such as
the peacekeeper missile fourth stage, the rocket engine developer has developed not only the engine,
but also much of the vehicle stage with its propellant tanks and pressurization system.

LIQUID PROPELLANT ROCKET ENGINE EUNDAMENTALS 197

TABLE 6-1. Characteristics of Two Categories of Liquid Propellant Rocket Engines

Purpose

Boost Propulsion

Auxiliary Propulsion

Mission

Impart significant velocity to

Attitude control, minor space

propel a vehicle along its

maneuvers, trajectory

flight path

corrections, orbit maintenance

Applications

Booster stage and upper stages

Spacecraft, satellites, top stage

of launch vehicles, large

of antiballistic missile, space

missiles

rendezvous

Total impulse

High

Low

Number of thrust

Usually 1; sometimes 4, 3, or 2

Between 4 and 24

chambers per
engine

Thrust level

High; 4500 N up to 7,900,000 N

Small; 0.001 up to 4500 N, a

or 1000-1,770,000 Ibf

few go up to 1000 Ibf

Feed system.

Mostly turbopump type;

Pressurized feed system with

typical

occasionally pressurized feed

high-pressure gas supply

Tank pressure

system for smaller thrusts
0.138-0.379 MPa or 20-55 psi

0.689-17.23 MPa or 100-2500

range

psi

Most common

Propellant cooled

Radiation cooled

cooling method

Propellants (see

Cryogenic and storable liquids

Storable liquids,

next section)

monopropellants, and/or stored

Chamber pressure

2.4-21 MPa or 350-3600 psi

cold gas

0.14-2.1 MPa or 20-300 psi

Number of starts

Usually no restart; sometimes

Several thousand starts are

during a single

one, but up to four in some

typical for small thrusters

mission

cases

Cumulative

Up to a few minutes

Up to several hours

duration of
firing

Shortest firing

Typically 5-40 sec

0.02 sec typical for small

duration

thrusters

Time elapsed to

Up to several seconds

Usually very fast, 0.004-0.080

reach full thrust

sec

Life in space

Hours, days, or months

10 years or more in space

rocket engine systems can be classified in several other ways. They can he
reusable (like the Space Shuttle main engine or a rocket engine for quick ascent
or maneuvers of fighter aircraft) or suitable for a single flight only (as the
engines in other launch vehicles), and they can he restartable, like a reaction
control engine, or single firing, as in space launch vehicle boosters. They can
also be categorized by their propellants, application, or stage, such as an upper
stage or booster stage, their thrust level, and hy the feed system type (pressurized
or turbopump).

198 LIQUID PROPELLANT ROCKET ENGINE FUNDAMENTALS

The thrust chamber or thruster is the combustion device where the liquid pro-
pellants are metered, injected, atomized, mixed, and burned to form hot gaseous
reaction products, which in turn are accelerated and ejected at a high velocity
to impart a thrust force. A thrust chamber has three major parts: an injector,
a combustion chamber, and a nozzle. In a cooled thrust chamber, one of the
propellants (usually the fuel) is circulated through cooling jackets or a special
cooling passage to absorb the heat that is transferred from the hot reaction gases
to the thrust chamber walls (see Figs. 8-2 and 8-9). A radiation-cooled thmst
chamber uses a special high-temperature material, such as niobium metal, which
can radiate away its excess heat. There are uncooled or heat-absorbing thmst
chambers, such as those using ablative materials. Thrust chambers are discussed
in Chapter 8.

There are two types of feed systems used for liquid propellant rocket engines:
those that use pumps for moving the propellants from their flight vehicle tanks to
the thmst chamber and those that use high-pressure gas for expelling or displacing
their propellants from their tanks. They are discussed further in Chapter 10 and
in Section 6.3.

Tables 19-1 to 19-4 compare the advantages and disadvantages of liquid
propellant rocket engines and solid propellant rocket motors.

6.1. TYPES OF PROPELLANTS

The liquid propellants, which are the working substance of rocket engines, con-
stitute the fluid that undergoes chemical and thermodynamic changes. The term
liquid propellant embraces all the various liquids used and may be one of the
following:

1. Oxidizer (liquid oxygen, nitric acid, nitrogen tetroxide, etc.)

2. Fuel (kerosene, alcohol, liquid hydrogen, etc.)

3. Chemical compound or mixture of oxidizer and fuel ingredients, capable
of self-decomposition, such as hydrazine

4. Any of the above, but with a gelling agent. Used only rarely.

All are described in Chapter 7.

A bipropellant has two separate liquid propellants, an oxidizer and a fuel. They
are the most common type. They are stored separately and are not mixed outside
the combustion chamber. A hypergolic bipropellant combination is self-igniting
upon contact between the oxidizer and the fuel. A nonhypergolic bipropellant
combination needs energy to start its combustion (e.g., heat or electric discharge)
and its engine needs an ignition system.

6.1 . TYPES OF PROPELLANTS 1 99

A monopropellant contains an oxidizing agent and combustible matter in a
single substance. It may be a mixture of several compounds or it may be a
homogeneous material, such as hydrogen peroxide or hydrazine. Monopropel-
lants are stable at ordinary atmospheric conditions but decompose and yield hot
combustion gases when heated or catalyzed.

A cold gas propellant (e.g., gaseous nitrogen or air) is stored at ambient
temperature but at very high pressure, gives a low performance, allows a simple
system, and is usually very reliable. It has been used for roll control and attitude
control.

A cryogenic propellant is liquified gas at low temperature, such as liquid oxy-
gen (— 183°C) or liquid hydrogen (— 253°C). Provisions for venting the storage
tank and minimizing vaporization losses are necessary with this type.

Storable propellants (e.g., nitric acid or gasoline) are liquid at ambient temper-
ature and at modest pressure and can be stored for long periods in sealed tanks.
Space- storable propellants are liquid in the environment of space; this storability
depends on the specific tank design, thermal conditions, and tank pressure. An
example is ammonia.

A gelled propellant is a thixotropic liquid with a gelling additive. It behaves
like a jelly or thick paint. It will not spill or leak readily, can flow under pressure,
will burn, and is safer in some respects. It is described at the end of Chapter 7.
Gelled propellants have been used in a few experimental rocket engines.

A hybrid propellant usually has a liquid oxidizer and a solid fuel. It is dis-
cussed in Chapter 16.

The propellant mixture ratio for a bipropellant is the ratio at which the oxidizer
and fuel are mixed and react to give hot gases. The mixture ratio r is defined as
the ratio of the oxidizer mass flow rate m„ and the fuel mass flow rate htf or

r = hiolhif (6-1)

As explained in Chapter 5 the mixture ratio defines the composition and tem-
perature of the combustion or reaction products. It is usually chosen to give a
maximum value of specific impulse or Ti/OTl, where T\ is the absolute combus-
tion temperature and Tl is the average molecular mass of the reaction gases (see
Eq. 3-16 or Fig. 3-2). For a given thrust F and a given effective exhaust veloc-
ity c, the total propellant flow is given by Eq. 2-6; namely, m = vv/go = F /c.
The relationships between r, rh, tko, and thf are

tho -\- thf = til

(6-2)

tho = rhijlr 1)

(6-3)

thf = m /(r -|- 1)

(6-4)

These same four equations are valid when w and vv (weight and weight flow
rate) are substituted for m and hi. Calculated performance values for a number

200 LIQUID PROPELLANT ROCKET ENGINE FUNDAMENTALS

of different propellant combinations are given for specific mixture ratios in Table
5-5. Physical properties and a discussion of several common liquid propellants
and their safety concerns are described in Chapter 7.

Example 6-1. A liquid oxygen-liquid hydrogen rocket thrust chamber of 10,000-lbf
thrust operates at a chamber pressure of 1000 psia, a mixture ratio of 3.40, has exhaust
products with a mean molecular mass of 8.90 Ibm/lb-mol, a combustion temperature of
4380°F, and a specific heat ratio of 1.26. Determine the nozzle throat area, nozzle exit area
for optimum operation at an altitude where = p 2 = 1.58 psia, the propellant weight
and volume flow rates, and the total propellant requirements for 2.5 min of operation.
Assume that the actual specific impulse is 97% of the theoretical value and that the thrust
coefficient is 98% of ideal value for this problem.

SOLUTION. The exhaust velocity for an optimum nozzle is determined from Eq. 3-16,
but with a correction factor of go for the foot-pound system:

2gpk R'Ti

k-l m

2 X 32.2 X 1.26 1544 x 4840

0.26

8.9

(1 - 0 . 001580 - 205 )

13,890 ft/sec

The theoretical specific impulse is c/go, or in this optimum expansion case V 2 /go or
13,890/32.2 = 431 sec. The actual specific impulse is 0.97 x 431 = 418 sec. The theo-
retical or ideal thrust coefficient can be found from Eq. 3-30 or from Fig. 3-6 (p 2 = pa)
and for a pressure ratio pi/p 2 = 633 to be Cf = 1.76. The actual thrust coefficient is
slightly less, say 98% or Cf = 1.72. The throat area required is found from Eq. 3-31.

A, = F/iCfPx) = 10,000/(1.72 x 1000) = 5.80 in.^ (2.71 in. diameter)

The optimum area ratio can be found from Eq. 3-25 or Eig. 3-5 to be 42. The exit area
is 5.80 X 42 = 244 in.^ (17.6 in. diameter). The weight density of oxygen is 71.1 Ibf/ft^
and of hydrogen is 4.4 Ibf/ft^. The propellant weight flow rates (Eqs. 2-5, 6-3, and 6-4)
are

w = F/E = 10,000/418 = 24.0 Ibf/sec

= wr/{r -H 1) = 24.0 x 3.40/4.40 = 18.55 Ibf/sec
Wf = w/{r -H 1) = 24/4.40 = 5.45 Ibf/sec

The volume flow rates are determined from the densities and the weight flow rates:

V„ = Wo/Po = 18.55/71.1 = 0.261 ftVsec
Vf = Wf/pf = 5.45/4.4 = 1.24 ftVsec

6.2. PROPELLANT TANKS 201

For 150 sec of operations (arbitrarily allow the equivalent of two additional seconds
for start and stop transients and unavailable residual propellant), the weight and volume
of required propellant are

Wo = 18.55 X 152 = 2820 Ibf of oxygen
w/ = 5.45 X 152 = 828 Ibf of hydrogen
Vo = 0.261 X 152 = 39.7 ft^ of oxygen
Vf = 1.24 X 152 = 188.5 ft^ of hydrogen

Note that, with the low-density fuel, the volume flow rate and therefore the tank volume
of hydrogen are 5 times as large compared to the oxidizer. Hydrogen is unique and in its
liquid state has a very low density.

6.2. PROPELLANT TANKS

In liquid bipropellant rocket engine systems the propellants are usually stored in
an oxidizer tank and a fuel tank within the flying vehicle. Monopropellant rocket
engine systems have, of course, only one propellant tank. There are usually
also one or more high-pressure gas tanks, the gas being used to pressurize the
propellant tanks. However, as will be discussed in Section 6.5, there are tank
pressurization schemes using heated gas from the engine without the need for
a high-pressure heavy gas storage tank. Tanks can be arranged in a variety of
ways, and the tank design, shape, and location can be used to exercise some
control over the change in the location of the vehicle’s center of gravity. Typical
arrangements are shown in Fig. 6-2. Because the propellant tank has to fly,
its mass is at a premium and the tank material is therefore highly stressed.
Common tank materials are aluminum, stainless steel, titanium, alloy steel, and
fiber-reinforced plastics with an impervious thin inner liner of metal to prevent
leakage through the pores of the fiber-reinforced waffs. Chapter 8 of Ref. 6-f
describes the design of propellant tanks.

The extra volume of gas above the propellant in sealed tanks is called ullage.
It is necessary space that allows for thermal expansion of the propellant liquids,
for the accumulation of gases that were originally dissolved in the propellant,
or for gaseous products from slow reactions within the propellant during stor-
age. Depending on the storage temperature range, the propellants’ coefficient of
thermal expansion, and the particular application, the ullage volume is usually
between 3 and 10% of the tank volume. Once propellant is loaded into a tank,
the ullage volume (and, if it is sealed, also its pressure) will change as the bulk
temperature of the propellant varies.

The expulsion efficiency of a tank and/or propellant piping system is the
amount of propellant expelled or available for propulsion divided by the total
amount of propellant initially present. Typical values are 97 to 99.7%. The losses

202 LIQUID PROPELLANT ROCKET ENGINE EUNDAMENTALS

Spherical tanks

Tandem tanks,
external piping

Tandem tanks
with common
bulkhead,
internal piping

3 Oxidizer

FIGURE 6-2. Simplified sketches of typical tank arrangements for large turbopump-fed
liquid propellant rocket engines.

are unavailable propellants that are trapped in grooves or corners of pipes, fit-
tings, and valves, are wetting the walls, retained by surface tension, or caught
in instrument taps. This residual propellant is not available for combustion and
must be treated as inert mass, causing the vehicle mass ratio to decrease slightly.
In the design of tanks and piping systems, an effort is made to minimize the
residual propellant.

The optimum shape of a propellant tank (and also a gas pressurizing tank) is
spherical because for a given volume it results in a tank with the least weight.
Small spherical tanks are often used with reaction control engine systems, where
they can be packaged with other vehicle equipment. Unfortunately, the larger
spheres, which are needed for the principal propulsion systems, are not very
efficient for using the space in a flight vehicle. These larger tanks are often made
integral with the vehicle fuselage or wing. Most are cylindrical with half ellipses
at the ends, but they can be irregular in shape. A more detailed discussion of
tank pressurization is given in Section 6.5.

Cryogenic propellants cool the tank wall temperature far below the ambient
air temperature. This causes condensation of moisture from the air on the outside
of the tank and usually also formation of ice during the period prior to launch.
The ice is undesirable because it increases the vehicle inert mass. Also, as pieces

6.2. PROPELLANT TANKS 203

of ice are shaken off or break off during the initial flight, these pieces can damage
the vehicle; for example, the ice from the Shuttle’s cryogenic tank can hit the
orbiter vehicle.

For an extended storage period, cryogenic tanks are usually thermally insu-
lated; porous external insulation layers have to be sealed to prevent moisture from
being condensed inside the insulation layer. With liquid hydrogen it is possible
to liquify or solidify the ambient air on the outside of the fuel tank. Even with
heavy insulation and low-conductivity structural tank supports, it is not possible
to prevent the continuous evaporation of the cryogenic fluid. Even with good
thermal insulation, all cryogenic propellants evaporate slowly during storage and
therefore cannot be kept in a vehicle for more than perhaps a week without
rehlling of the tanks. Eor vehicles that need to be stored or to operate for longer
periods, a storable propellant combination is preferred.

Prior to loading very cold cryogenic propellant into a flight tank, it is necessary
to remove or evacuate the air to avoid forming solid air particles or condensing
any moisture as ice. These frozen particles would plug up injection holes, cause
valves to freeze shut, or prevent valves from being fully closed. Tanks, piping, and
valves need to be chilled or cooled down before they can contain cryogenic liquid
without excessive bubbling. This is usually done by letting an initial amount of
cryogenic liquid absorb the heat from the relatively warm hardware prior to
engine start. This initial cool-down propellant is vaporized and vented through
appropriate vent valves and cannot be used for propulsion.

If the tank or any segment of piping containing low-temperature cryogenic
liquid is sealed for an extended period of time, heat from ambient-temperature
hardware will result in evaporation, and this will greatly raise the pressure
until it exceeds the strength of the container (see Ref. 6-3). Controlled self-
pressurization can be difficult to achieve. Uncontrolled self-pressurization will
cause a failure, usually a major leak or even a tank explosion. All cryogenic
tanks and piping systems are therefore vented during storage on the launch pad,
equipped with pressure safety devices (such as burst diaphragms or relief valves),
and the evaporated propellant is allowed to escape from its container. Eor long-
term storage of cryogenic propellants in space vacuum (or on the ground) some
form of a powered refrigeration system is needed to recondense the vapors and
minimize evaporation losses. The tanks are usually rehlled or topped off just
before launch to replace the evaporated and vented cool-down propellant. When
the tank is pressurized, just before launch, the boiling point is usually raised
slightly and the cryogenic liquid can usually absorb the heat transferred to it
during the several minutes of rocket firing.

There are several categories of tanks in liquid propellant propulsion systems.
There are a few exceptions to the pressure values listed below.

1 . Eor pressurized feed systems the propellant tanks typically operate at an
average pressure between 1.3 and 9 MPa or about 200 to 1800 Ibf/in.^
These tanks have thick walls and are heavy.

204 LIQUID PROPELLANT ROCKET ENGINE FUNDAMENTALS

2. For high-pressure gas (used to expel the propellants) the tank pressures are
much higher, typically between 6.9 and 69 MPa or 1000 to 10,000 Ibf/in.^.
These tanks are usually spherical for minimum inert mass. Several small
spherical tanks can be connected together, and then they are relatively easy
to place within the confined space of a vehicle.

3. For turbopump feed systems it is necessary to pressurize the propellant
tanks slightly (to suppress pump cavitation as explained in Sections 10.3
and 10.4) to average values of between 0.07 and 0.34 MPa or 10 to 50
Ibf/in^. These low pressures allow thin tank walls, and therefore turbopump
feed systems have relatively low inert tank weights.

Liquid propellant tanks can be difficult to empty under side accelerations,
zero-g, or negative-g conditions during flight. Special devices and special types
of tanks are needed to operate under these conditions. Some of the effects that
have to be overcome are described below.

The oscillations and side accelerations of vehicles in flight can cause sloshing
of the liquid in the tank, very similar to a glass of water that is being jiggled. In
an anti-aircraft missile, for example, the side accelerations can be large and can
initiate severe sloshing. Typical analysis of sloshing can be found in Refs. 6-4
and 6-5. When the tank is partly empty, sloshing can uncover the tank outlet
and allow gas bubbles to enter into the propellant discharge line. These bubbles
can cause major combustion problems in the thrust chambers; the aspirating
of bubbles or the uncovering of tank outlets by liquids therefore needs to be
avoided. Sloshing also causes shifts in the vehicle’s center of gravity and makes
flight control difficult.

Vortexing can also allow gas to enter the tank outlet pipe; this phenomenon
is similar to the Coriolis force effects in bath tubs being emptied and can be
augmented if the vehicle spins or rotates in fight. Typically, a series of internal
baffles is often used to reduce the magnitude of sloshing and vortexing in tanks
with modest side accelerations. A positive expulsion mechanism can prevent gas
from entering the propellant piping under multidirectional major accelerations
or spinning (centrifugal) acceleration. Both the vortexing and sloshing can also
greatly increase the unavailable or residual propellant, and thus cause a reduction
in vehicle performance.

In the gravity-free environment of space, the stored liquid will float around
in a partly emptied tank and may not always cover the tank outlet, thus allow-
ing gas to enter the tank outlet or discharge pipe. Figure 6-3 shows that gas
bubbles have no orientation. Various devices have been developed to solve this
problem: namely, positive expulsion devices and surface tension devices. The
positive expulsion tank design includes movable pistons, inflatable flexible blad-
ders, or thin movable, flexible metal diaphragms. Surface tension devices rely on
surface tension forces to keep the outlet covered with liquid.

Several basic types of positive expulsion devices have been used successfully
in propellant tanks of pressurized feed systems. They are compared in Table 6-2

6.2. PROPELLANT TANKS 205

Liquid out

Outlet

tension propellant
management device
showing one particular
liquid distribution during
upward acceleration

in a zero-gravity environment; surface
liquid.

FIGURE 6-3. Ullage bubbles can float around
tension device can keep tank outlet covered with

and shown in Fig. 6-4 for simple tanks. These devices mechanically separate
the pressurizing gas from the liquid propellant in the propellant tank. Separation
is useful for these reasons:

1 . It prevents pressurizing gas from dissolving in the propellant and propellant
from evaporating. Dissolved pressurizing gas dilutes the propellant, reduces
its density as well as its specific impulse, and makes the pressurization
inefficient.

2. It allows moderately hot and reactive gases (such as gas generated by gas
generators) to be used for pressurization, and this permits a reduction in
pressurizing system mass and volume. The mechanical separation prevents
a chemical reaction between the hot gas and the propellant, prevents gas
from being dissolved in the propellant, and reduces the heat transfer to the
liquid.

3. In some cases tanks containing toxic propellant must be vented without
spilling any toxic liquid propellant or its vapor. For example, in servicing
a reusable rocket, the tank pressure needs to be relieved without venting
or spilling potentially hazardous material.

TABLE 6-2. Comparison of Propellant Expulsion Methods for Spacecraft Hydrazine Tanks

<D C ^

^ .2 S

c h

c/5 H

° ^
° s

tc D
£ CQ

— ^
- .2 Si'S
-S S ^ G.

^ Oj Oh ^

£ s s 1

I D

== .2
^ P QJ

D S "O
X P ^

izi --r
^ «
w

■y u

S H

0J5 P ^

C O -p Oh

_ .4-1 D . cn

^ J S E

cq Q u

O 2 25

> 73

IZ2 D

g 8 'S

^ c >

1-1

d a j

o a

too a
>» •B

’ o

t 3 ;2
^ o “

o S

OJ >

P X o

-H W J

■T^ ^

" c

s .§

■y '3 o

s S :s

S § S

13 xi
^ ^ a

-H W ffi

> IZ3

^ O

o o

^ -2
o ^

o a a
X o .a
PJ z J

Vh >

^ 2 •
-^-T g bJ5

■3 .S c

X, ■■^ ,i=> CL) "O

^ 2 13 =a o

s o T3 C3 .G

9 « ^
O J ^

t3 __

8 tS 3 _

60 4= '3 _3
>4-3 G C6

Lh Cd -4-* <H-<

X D O

> J Oh

J3 'C Cfl
w CJ (U (L)

O >

.2 S

CO JC

° c

c/3 ^

c ^ X

^ 2 Oh i£, ^

"u G3 S u

^ ^ -O cj

w j S

o 3

S 60
S 'S

60 S

c/3

i-i

’c

c/3

M-h

-1

C ^

W J u

*C/3

o >,

D 8
2 -O
> o

H-.

w t-

a
>

OO g ID !«

j ( i : S

206

6.2. PROPELLANT TANKS 207

Sliding piston with seals

Propellant
outlet pipe

FIGURE 6-4. Three concepts of propellant tanks with positive expulsion: (a) inflatable
dual bladder; (b) rolling, peeling diaphragm; (c) sliding piston. As the propellant volume
expands or contracts with changes in ambient temperature, the piston or diaphragm will
also move slightly and the ullage volume will change during storage.

A piston expulsion device permits the center of gravity (CG) to be accurately
controlled and its location to be known during the engine operation. This is
important in rockets with high side accelerations such as anti-aircraft missiles or
space defense missiles, where the thrust vector needs to go through the vehicle’s
CG; if the CG is not well known, unpredictable turning moments may be imposed
on the vehicle. A piston also prevents sloshing or vortexing.

Surface tension devices use capillary attraction for supplying liquid propellant
to the tank outlet pipe. These devices (see Fig. 6-3) are often made of very
fine (300 mesh) stainless steel wire woven into a screen and formed into tunnels
or other shapes (see Refs. 6-6 and 6-7). These screens are located near the
tank outlet and, in some tanks, the tubular galleries are designed to connect
various parts of the tank volume to the outlet pipe sump. These devices work
best in a relatively low-acceleration environment, when surface tension forces
can overcome the inertia forces.

The combination of surface tension screens, baffles, sumps, and traps is called
a propellant management device. Although not shown in any detail, they are
included inside the propellant tanks of Figs. 6-3 and 6-13.

208 LIQUID PROPELLANT ROCKET ENGINE FUNDAMENTALS

High forces can be imposed on the tanks and thus on the vehicle by strong
sloshing motions of the liquid and also by sudden changes in position of liquid
mass in a partly empty tank during a gravity-free flight when suddenly accelerated
by a relatively large thrust. These forces can be large and can cause tank failure.
The forces will depend on the tank geometry, baffles, ullage volume, and its
initial location and the acceleration magnitude and direction.

6.3. PROPELLANT FEED SYSTEMS

The propellant feed system has two principal functions; to raise the pressure of
the propellants and to feed them at predicted mass flow rates to one or more thrust
chambers. The energy for these functions comes either from a high-pressure gas,
centrifugal pumps, or a combination of the two. The selection of a particular feed
system and its components is governed primarily by the application of the rocket,
the requirements mentioned at the beginning of this chapter, duration, number or
type of thrust chambers, past experience, mission, and by general requirements
of simplicity of design, ease of manufacture, low cost, and minimum inert mass.
A classification of several of the more important types of feed system is shown
in Fig. 6-5, and some are discussed in more detail in other parts of this book. All
feed systems have piping, a series of valves, provisions for filling and usually also

Liquid propellant
feed systems

Pressurized systems

Turbopump systems

Direct gas
pressurization

Flexible bag
within tank

Piston

pressurization

Pump

By stored

By vaporized

By chemically

inert gas

propellant

generated gas

Multi-stage

Sinale staae

impellers

impeller

As received

Heated

Regulated

Blow down

pressure

Small portion of one
propellant flow raised to
precombustor pressure
with additional impeller

One pump
each for fuel
and oxidizer

Two main
pumps plus two
booster pumps

Gas power supply
and gas discharge

Precombustion

chamber

Separate gas
generator

Warm
hydrogen
from cooling
jacket

Driven by high pressure
liquid propellant (for
booster pumps only)

Hot gas bleed
from main
combustion
chamber

Flow through
turbines in
series

Flow through
turbines in
parallel

One turbine drives
both fuel and
oxidizer pumps

Two turbines: one
for fuel pump, one
for oxidizer pump

Four turbines; two
for main pumps, two
for booster pumps

Exhaust overboard

Exhaust feeds

Exhaust into

with low area

into diverging

injector of

ratio nozzle

nozzle section

main combustor

With gear

Direct drive

transmission

FIGURE 6-5. Design options of feed systems for liquid propellant rocket engines. The
more common types are designated with a double line at the bottom of the box.

6.3. PROPELLANT FEED SYSTEMS 209

for removing (draining and flushing) the liquid propellants, and control devices
to initiate, stop, and regulate their flow and operation. See Ref. 6-1.

In general, a gas pressure feed system gives a vehicle performance superior
to a turbopump system when the total impulse or the mass of propellant is
relatively low, the chamber pressure is low, the engine thrust-to-weight ratio is
low (usually less than 0.6), and when there are repeated short-duration thrust
pulses; the heavy- walled tanks for the propellant and the pressurizing gas usually
constitute the major inert mass of the engine system. In a turbopump feed system
the propellant tank pressures are much lower (by a factor of 10 to 40) and thus the
vehicle’s tank masses are much lower (again by a factor of 10 to 40). Turbopump
systems usually give a superior vehicle performance when the total impulse is
large (higher Am) and the chamber pressure is higher.

Local Pressures and Flows

Key parameters of any feed system and any liquid propellant rocket engine are
the flow magnitudes and the local pressures of the pressurizing gas subsystem,
the oxidizer, and the fuel flow subsystem including the flow passages of thrust
chambers. An inspection of the flow diagram of a relatively simple rocket engine
with a pressurized feed system, similar to Fig. 1-3, shows that the gas flow splits
into two branches, and the propellant flow splits into pipes leading to each of the
thrust chambers. The highest pressure is in the high-pressure gas supply tank. The
pressure drops in the pressurizing gas subsystem (pipes, valves, regulator) and
then drops further in the liquid propellant flow subsystems (more pipes, valves,
filters, injector, or cooling jacket) and the flows go into the thrust chamber, where
the propellants burn at chamber pressure. The pressure reaches a minimum at the
nozzle exit. If enough information is known about the geometric details and
functions of the components and their flow passages, then it has been possible
to use computer programs to obtain these pressure drops and flow distributions.
If this analysis can be validated by data from previous pertinent tests, it will be
more useful. Many rocket engine design organizations have developed their own
computer programs for estimating pressures and flows in different parts of an
engine.

Analyzing and knowing the local branch flows and local pressures is important
for the following reasons:

1 . They are used in the stress analysis and sometimes in the thermal analysis
of components and subsystems.

2. They are needed in the calibration of the rocket engine, so that this engine
will operate at the intended mixture ratio, chamber pressure, or thrust.
This is accomplished by using control devices or simple orifices to adjust
the pressures. This calibration also requires the proper balance of flows
and pressures. For a feed system with one or more turbopumps, such as
the one in Fig. 1-4, one needs to also include the rise in pressure as
propellants flow through a pump. Furthermore there is another combustion

210

LIQUID PROPELLANT ROCKET ENGINE FUNDAMENTALS

device (gas generator or prebumer) or alternatively the evaporation of a
cryogenic propellant in a cooling jacket, and this makes the feed system
with turbopumps more complex. A more detailed discussion of engine
calibration is given in Section 11.5.

3. Measuring the actual flows and several key local pressures during engine
ground tests or actual flight operation and then comparing these actual
values with the calculated or predicted values makes is possible to iden-
tify discrepancies between practice and theory. Such discrepancies give
clues to possible malfunction, inadequate design, or poor fabrication, which
can then be identified and probably corrected. If the real measured values
can be compared with the analysis values in real time, some experimental
test hardware can be saved from self-destruction. This can be the basis
for a real-time engine health monitoring system, which is discussed in
Section 11.4.

Some of the analyses are aimed at transient conditions, such as during the start
period, the shutdown period, or the change in thrust value (throttling). These tran-
sient analyses provide for the filling of empty propellant passages with propellant,
for different propellant temperatures, water hammer, valve reaction time, and the
like.

6.4. GAS PRESSURE FEED SYSTEMS

One of the simplest and most common means of pressurizing the liquid pro-
pellants is to force them out of their respective tanks by displacing them with
high-pressure gas. Rocket engines with pressurized gas feed systems can be very
reliable. References 6-1, 6-8, and 6-9 give additional information. A rocket
engine with a gas-pressurized feed system was the first to be tested and flown
(1926). Since then there have been two common types of pressurized feed sys-
tems and both are still used often today. The first uses a gas pressure regulator
in the gas feed line and its engine operates at essentially constant tank pressure
and nearly constant thrust.

It is shown schematically in Fig. 1-3 and consists of a high-pressure gas tank,
a gas starting valve, a pressure regulator, propellant tanks, propellant valves,
and feed lines. Additional components, such as filling and draining provisions,
check valves, filters, flexible elastic bladders for separating the liquid from the
pressurizing gas, and pressure sensors or gauges, are also often incorporated.
After all tanks are filled, the high-pressure gas valve in Fig. 1-3 is remotely
actuated and admits gas through the pressure regulator at a constant pressure to
the propellant tanks. The check valves prevent mixing of the oxidizer with the
fuel, particularly when the unit is not in an upright position. The propellants are
fed to the thrust chamber by opening appropriate valves. When the propellants
are completely consumed, the pressurizing gas can also scavenge and clean lines
and valves of much of the liquid propellant residue. The variations in this system,
such as the combination of several valves into one or the elimination and addition

6.4. GAS PRESSURE EEED SYSTEMS 21 1

of certain components, depend to a large extent on the application. If a unit is to
be used and flown repeatedly, such as a space-maneuver rocket, it will include
several additional features such as, possibly, a thrust-regulating device and a tank
level gauge; they will not be found in an expendable, single-shot unit, which may
not even have a tank-drainage provision.

The second common type of gas pressure feed system is called a blow-down
feed system. It is shown in Fig. 6-6 and discussed in Ref. 6-10. Here the
propellant tanks are enlarged because they store not only the propellants but also
the pressurizing gas at the initial maximum propellant tank pressure. There is no
separate high-pressure gas tank and no pressure regulator. The expansion of the
gas already in the tanks provides for the expulsion of the propellants from their
tanks. The blow-down system can be lighter than a regulated pressure system, but
the thrust and the pressures decrease steadily as the propellants are consumed. A
comparison of these two common types is shown in Table 6-3.

Different bipropellant pressurization concepts are evaluated in Refs. 6-1, 6-8,
6-9, and 6-10. Table 6-4 lists various optional features aimed at satisfying
particular design goals. Many of these features also apply to pump-fed systems,
which are discussed in Section 6.6. Many feed systems or engines have some,
but certainly not all, of the features listed in Table 6-4. With monopropellants
the gas pressure feed system becomes simpler since there is only one propellant
and not two, reducing the number of pipes, valves, and tanks.

FIGURE 6-6. Simplified schematic diagram of a bipropellant blow-down pressurized
gas feed system with two thrusters.

212 LIQUID PROPELLANT ROCKET ENGINE FUNDAMENTALS

TABLE 6-3. Comparison of Two Types of Gas Pressurization Systems

Type

Regulated Pressure

Blowdown

Pressure/thrust

Stays essentially constant

Decreases as propellant is consumed

Gas storage

In separate high-pressure tanks

Gas is stored inside propellant tank
with large ullage volume
(30-60%)

Required

components

Needs regulator, filter, gas valve,
and gas tank

Larger, heavier propellant tanks

Advantages

Constant-pressure feed gives Simpler system. Less gas required

essentially constant propellant Can be less inert mass

flow and approximately constant
thrust, constant R and r
Better control of mixture ratio

Disadvantages

Slightly more complex
Regulator introduces a small
pressure drop

Gas stored under high pressure
Shorter burning time

Thrust decreases with burn duration
Somewhat higher residual
propellant due to less accurate
mixture ratio control
Thruster must operate and be stable
over wide range of thrust values
and modest range of mixture ratio
Propellants stored under pressure;
slightly lower R toward end of
burning time

An example of a complex man-rated pressurized feed system, the combined
Space Shuttle Orbital Maneuver System (OMS) and the Reaction Control Sys-
tem (RCS), is described in Figs. 6-7 and 6-8, and Ref. 6-11. There are three
locations on the Space Shuttle for the RCS, as shown in Fig. 1-14: a forward
pod and a right and left aft pod. Figures 6-7 and 6-8 and Table 6-5 refer to
one of the aft pods only and show a combined OMS and RCS arrangement.
The OMS provides thrust for orbit insertion, orbit circularization, orbit transfer,
rendezvous, deorbit, and abort. The RCS provides thrust for attitude control (in
pitch, yaw, and roll) and for small-vehicle velocity corrections or changes in
almost any direction (translation maneuvers), such as are needed for rendezvous
and docking; it can operate simultaneously with or separate from the OMS.

The systems feature various redundancies, an automatic RCS thruster selec-
tion system, various safety devices, automatic controls, sensors to allow a display
to the shuttle’s crew of the system’s status and health, and manual command
overrides. The reliability requirements are severe. Several key components, such
as all the helium pressure regulators, propellant tanks, some valves, and about
half the thrusters are duplicated and redundant; if one fails, another can still
complete the mission. It is possible to feed up to 1000 Ibm of the liquid from
the large OMS propellant tanks to the small RCS ones, in case it is neces-
sary to run one or more of the small reaction control thrusters for a longer

6.4. GAS PRESSURE FEED SYSTEMS 213

TABLE 6-4. Typical Features of Liquid Propellant Feed Systems

Enhance Safety

Check valves to prevent backflow of propellant into the gas tank and inadvertent mixing of
propllants inside flow passages

Pressurizing gas should be inert, clean, and insoluble in propellant

Burst diaphragms or isolation valves to isolate the propellants in their tanks and positively prevent
leakage into the thrust chamber or into the other propellant tank during storage
Isolation valves to shut off a section of a system that has a leak or malfunction
Sniff devices to detect leak of hazardous vapor; used on some versions of Space Shuttle Orbiter
Features that prevent an unsafe condition to occur or persist and shut down engine safely, such as
relief valves or relief burst diaphragms to prevent tank overpressurization), or a vibration
monitor to shut off operation in the case of combustion instability

Provide Control

Valves to control pressurization and flow to the thrust chambers (start/stop/throttle)

Sensors to measure temperatures, pressures, valve positions, thrust, etc., and computers to
monitor/analyze system status. Compare measured values with analytical estimates, issue
command signals, and comect if sensed condition is outside predetermined limits
Manned vehicle can require system status display and command signal override
Fault detection, identification, and automatic remedy, such as shut-off isolation valves in
compartment in case of fire, leak, or disabled thruster
Control thrust (throttle valve) to fit a desired thrust-time flight profile

Enhance Reliability

Fewest practical number of components/subassemblies

Filters to catch dirt in propellant lines, which could prevent valve from closing or small injector
holes from being plugged up or bearings from galling
Duplication of key components, such as redundant small thrusters, regulators, or check valves. If
malfunction is sensed, then remedial action is to use the redundant spare component
Heaters to prevent freezing of moisture or low-melting-point propellant

Long storage life — use propellants with little or no chemical deterioration and no reaction with
wall materials

Provide for Reusability

Provisions to drain propellants or pressurants remaining after operation

Provision for cleaning, purging, flushing, and drying the feed system and refilling propellants and
pressurizing gas

Devices to check functioning of key components prior to next operation
Features to allow checking of engine calibration and leak testing after operation
Access for inspection devices for visual inspection of internal surfaces or components for damage
or failure

Enable Effective Propellant Utilization

High tank expulsion efficiency with minimum residual, unavailable propellant
Lowest possible ambient temperature variation and/or matched propellant property variation with
temperature so as to minimize mixture ratio change and residual propellant
Alternatively, measure remaining propellant in tanks (using a special gauge) and automatically
adjust mixture ratio (throttling) to minimize residual propellant
Minimize pockets in the piping and valves that cannot be readily drained or flushed

214 LIQUID PROPELLANT ROCKET ENGINE EUNDAMENTALS

RCS helium RCS propellant Gimballed

tanks manifold valves OMS engine

FIGURE 6-7. Simplified sketch at the left aft pod of the Space Shuttle’s Orbiting Maneu-
vering System (OMS) and the Reaction Control System (RCS). (Source: NASA.)

period and use more propellant than the smaller tanks allow; it is also possible
to feed propellant from the left aft system to the one on the vehicle’s right side,
and vice versa. These features allow for more than nominal total impulse in a
portion of the thrusters, in case it is needed for a particular mission mode or an
emergency mode.

The compartmented steel propellant tanks with antislosh and antivortex baf-
fles, sumps, and a surface tension propellant retention device allow propellant
to be delivered independent of the propellant load, the orientation, or the accel-
eration environment (some of the time in zero-g). Gauges in each tank allow a
determination of the amount of propellant remaining, and they also indicate a
leak. Safety features include sniff lines at each propellant valve actuator to sense
leakage. Electrical heaters are provided at propellant valves, certain lines, and
injectors to prevent fuel freezing or moisture forming into ice.

Some pressure feed systems can be prefilled with storable propellant and pres-
surizing agent at the factory and stored in readiness for operation. Compared to
a solid propellant rocket unit, these storable prepackaged liquid propellant pres-
surized feed systems offer advantages in long-term storability, random restarts,
and resistance to transportation vibration or shock.

6.4. GAS PRESSURE FEED SYSTEMS 215

TABLE 6-5. Characteristics of the Orbital Maneuver System (OMS) and the Reaction
Control System (RCS) of the Space Shuttle in one of the Aft Side Pods

Item

OMS

Primary RCS

Vernier RCS

Thrust (per nozzle) (Ibf)

6000

870

Number of thrusters per pod

Thrust chamber cooling

Regenerative and radiation

Radiation cooling

Chamber pressure, nominal

125

152

110

(psi)

Specific impulse (vacuum

313

280“

265“

nominal) (sec)

Nozzle area ratio

22-30"

20-50“

Mixture ratio (oxide/fuel mass

1.65

1.6

flow)

Bum time, minimum (sec)

0.08

Bum time, maximum (sec)

160

150

125

Bum time, cumulative (sec)

54,000

12,800

125,000

Number of starts, cumulative

1000

20,000

330,000

(sec)

Oxidizer (N 2 O 4 ) weight in

14,866

1464

tank (lb)

Fuel (MMH) weight in tank

9010

923

(lb)

Number of oxidizer/fuel tanks

1/1

Propellant tank volume, each

17.9

tank (ft^)

Ullage volume, nominal (full

7.8

1.2

-1.5

tank) (ft^)

Tank pressure, nominal (psi)

250

280

Helium storage tank pressure

4700

3600

(psi)

Number of helium tanks

Volume of helium tanks (ft^)

1.76

“Depends on specific vehicle location and scarfing of nozzle.
Sources: NASA, Aerojet General Corporation.

The thrust level of a rocket propulsion system with a pressurized gas feed
system is determined by the magnitude of the propellant flows which, in turn,
is determined by the gas pressure regulator setting. The propellant mixture ratio
in this type of feed system is controlled by the hydraulic resistance of the liq-
uid propellant lines, cooling jacket, and injector, and can usually be adjusted by
means of variable restrictors or fixed orifices. Further discussion of the adjust-
ing of thrust and mixture ratio can be found in Sections 11.5 and 11.6 and in
Example 11-2.

- 5 5 o 4

- E 2 ^

-o o

C C3
N

2 -S

-5

0
d

V 5

too

0 ) d)

> c

D ^

d c^
.■d T3
X) . ..

U /—N

°d

■£ Z

^ CD

^ "S

^ n

a; eu 3

^ P

l-H . .

0 !Z 3

tM (U

!C ^

1 ^
V 5 O

T 3 S

,D >

C .-d
d ^
X Vh
D O

o ii

U ti

G, S
D X
X C/5

^ “

® d

u lU

bJ) X
ra •'“'

0 ^

X ^

X w

^ r-

X £
X D

£ ^
•P c/5
00 ^
O

00 ^

1 ^

^ rS

216

(MMH); short dashed lines: high-pressure helium. (Source: NASA.)

6.5. TANK PRESSURIZATION 217

6.5. TANK PRESSURIZATION

As was stated before the objective of feed systems is to move the propellants
under pressure from the propellant tanks to the thrust chamber(s). The tank pres-
surization system is that part of the feed system that provides a gas to expel
propellants from the tanks. See Refs. 6-1, 6-8, 6-9, and 6-10. As was described
in Section 6.3 there are two types: (1) in a pressurized gas feed system a rel-
atively high pressure gas displaces the propellants from the tanks and (2) in a
pumped feed system (described in the next section) the main energy for feeding
the propellants comes from one or more pumps, but it requires a low gas pressure
in the tanks to move the propellants to the pump inlet in order to avoid pump
cavitation.

There are several sources of pressurizing gas commonly used in tank pressur-
ization systems.

1 . High-pressure inert gas stored at ambient temperature is the most common.
Typical gases are helium, nitrogen, or air. Table 6-3 shows a comparison
of the regulated pressure system (see Fig. 1-3) and the blow-down system
(see Fig. 6-6). It is discussed further in this section. As the gas expands,
its temperature drops.

2. Heated high-pressure inert gas (typically 200 to 800°F or 93 to 427°C)
reduces the amount of required gas mass and thus the inert mass of the
pressurizing system. Examples are gas heated by a heat exchanger with
warm gas from a gas generator, with warm gas from turbine exhaust, or
with electrical heaters inside the gas tank. Reference 6-12.

3. Gas is created by a chemical reaction by using either liquid bipropellants
or a monopropellant and alternatively by a solid propellant, all at mixture
ratios that allow a low enough or “warm” gas reaction temperature, so
uncooled hardware can be used. The term “warm gas” (say 400 to 1600°F
or 204 to 871°C) is used here to distinguish it from the hot gas (4000 to
6000°F, or 2204 to 3319°C) in the main combustion chamber. This chem-
ically generated warm gas usually results in a lighter tank pressurization
system than the heated inert gas system, particularly for applications with a
high total impulse. This gas can come from two separate small gas genera-
tors for tank pressurization; one produces fuel-rich gas for pressurizing the
fuel tank and the other feeds oxidizer-rich pressurizing gas into the oxidizer
tank. This scheme was first used in the United States on the Bomarc rocket
engine around 1952 and on the Russian RD-253 around 1961. The other
common warm-gas scheme is to bleed a small amount of warm gas from the
engine’s gas generator or preburner. If this gas is fuel rich, then it can only
be used for pressurizing the fuel tank, and it may need to be cooled. These
chemical reaction gases typically are at 1200 to 1700°F or 649 to 921°C,
which are within the gas temperatures allowable for most alloy turbines and
steel tanks. Catalyzed decomposed hydrazine warm gaseous reaction prod-
ucts have successfully pressurized liquid hydrazine tanks. With aluminum

218

LIQUID PROPELLANT ROCKET ENGINE FUNDAMENTALS

propellant tanks it is necessary to cool the warm gas further. Many alu-
minum alloys melt around 1100°F (593°C). This cooling of gas has been
accomplished by using a heat exchanger with one of the propellants. Solid
propellant gas generators have been used on experimental liquid propellant
rocket engines, but as far as the author can determine, none have been
adapted for a production flight vehicle. One clever system developed in
the former Soviet Union uses two rocket engines operating simultaneously
on the same vehicle. The larger engine has a bleed of oxidixer-rich gases
from an oxidizer-rich preburner for pressurizing the common oxidizer tank.
The second engine feeds four smaller hinge-mounted vernier thrust cham-
bers, which hare been used for attitude control and also for extra thrust;
its fuel-rich gas generator has a bleed of gas for pressurizing the common
fuel tank. See Ref. 6-12.

4. Evaporate a flow of a cryogenic liquid propellant , usually liquid hydrogen,
by applying heat usually taken from a thrust chamber cooling jacket and
then use a part or all of this evaporated flow for tank pressurization. Orifices
or pressure regulators may be needed to control the desired tank pressure.
This scheme is used for the fuel tank of the Space Shuttle Main Engine and
the tap-off stub for pressurizing gas can be seen in Fig. 6-12 in the turbine
exhaust manifold of the fuel booster pump. The oxygen tank is pressurized
by gasified liquid oxygen, which was tapped off the discharge side of the
main oxygen pump and heated in a heat exchanger around the turbine of
the main oxygen pump, as shown in Figure 6-12. See Ref. 6-12.

5. Direct injection of a small stream of hypergolic fuel into the main oxidizer
tank and a small flow of oxidizer into the fuel tank has been tried in several
countries without success. It is really a form of chemical gas generation.
The only successful flight application is reported to be with a Russian
ICBM upper stage engine using hypergolic propellants. See Refs. 6-1 and
6 - 12 .

6. Self-pressurization of cryogenic propellants by evaporation is feasible but
can be difficult to control (see Ref. 6-3). Experience is limited.

In order to design or analyze a pressurizing system it is necessary to know
or to assume certain data about the tank and the engine. This can include the
basic engine parameters, such as the propellant flow, thrust, duration, pulse width,
propellant tank volume, percent ullage of that volume, storage temperature range,
propellant and pressurizing gas properties, propellant tank pressure, gas tank
pressure, or amount of unavailable residual propellant. Eor many of these the
nominal, maximum, and minimum values may be needed.

Factors Influening the Required Mass of Pressurizing Gas

A key task in the analysis and design of tank pressurization systems is determining
the required mass of pressurizing gas. Many different factors influence this gas
mass, and some of them can be quite complex as shown in Ref. 6-1. Therefore

6.5. TANK PRESSURIZATION 219

a basic analysis of the amount of pressurizing gas for a new propulsion system
is usually not accurate or fully useful. The mass of gas can be estimated using
simplifying assumptions, but it is more accurate if it is based on actual test
results and/or data from similar proven pressurizing systems. These are some of
the influencing factors:

1 . Evaporation of propellant at the interface between the pressurizing gas
and the liquid propellant is a key phenomenon. The evaporated propellant dilutes
the gas and changes its properties and expansion behavior. It depends on the
temperature difference between the gas and the liquid, sloshing, vapor pressure
of the propellant, turbulence, and local velocities of the gas. Furthermore the
him of propellant on those portions of the tank walls and baffles, which are
above the liquid level, will also evaporate. “Warm” gases (e.g., a bleed from a
gas generator) will cause heating of the top layer of the liquid propellant and
increase propellant evaporation. With cryogenic propellants the gas is always
warmer than the liquid. The gas is cooled and more liquid evaporates.

2. The temperature of those walls of the propellant tanks, which are part of the
exterior vehicle surface, can be affected by aerodynamic heating, which varies
during the flight. This heat can increase the gas temperature as well as the liquid
propellant temperature, and can augment the liquid evaporation.

3. The solubility of the gas in the liquid is affected by temperature and pres-
sure. For example, nitrogen gas is soluble in liquid oxygen. It requires four times
as much nitrogen gas to pressurize liquid oxygen as an equivalent volume of
water. This dilutes the oxygen and causes a small performance loss. Helium is
not very soluble in oxygen.

4. Condensation of certain species of the gas can dilute the propellant. Water
vapor in warm gases is an example. Condensation can also occur on the exposed
wetted inner walls of the propellant tanks and this requires more pressurizing
gas.

5. Changes in the gas temperature take place during the operation. The com-
pressed gas undergoes an expansion that can cause a dramatic cooling of the gas.
Temperatures as low as 160 to 200 K (—228 to — 100°F) have been recorded
with helium. The cold gas absorbs heat from the propellants and the engine
hardware. The particular expansion process will depend largely on the time of
rocket operation. For large liquid propellant rocket engines, which run only for
a few minutes, the expansion process will be close to adiabatic, which means
little heat transfer from the hardware to the gas. For satellites, which stay in
orbits for years and where the thrusters operate only occasionally for short oper-
ating periods, heat will be transferred from the vehicle hardware to the gas
and the process will be close to an isothermal expansion (no change in tempe-
rature).

6. Chemical reactions of some species of the pressurizing gas with the liquid
propellant have occurred and some can generate heat. Inert gases, such as helium,
have no chemical reactions with the propellants.

220 LIQUID PROPELLANT ROCKET ENGINE FUNDAMENTALS

7. Turbulence and irregular flow distribution of the entering gas will increase
the heat transfer between the liquid and the gas. Depending on the temperature
difference, it can cause additional heating or cooling of the top liquid layers.

8. Vigorous sloshing can quickly change the gas temperature. In some of
the experimental flights of the Bomarc missile the side accelerations induced
sloshing, which caused a sudden cooling of the warm tank pressurizing gas and
caused a sudden reduction of tank pressure and propellant flow. See Refs. 6-4
and 6-5.

9. In many rocket engines some of the pressurizing gas is used for purposes
other than tank pressurization, such as actuation of valves or controls. The amount
required has to he determined and added to the total needed gas mass.

Simplified Analysis for the Mass of Pressurizing Gas

This is one example for a pressurizing system using a compressed gas initially
in a separate tank at ambient temperature. It is the first category of pressurizing
systems discussed above and it is perhaps the most common type. It assumes an
adiabatic expansion of the gas (implies no heat transfer to or from the system
hardware), the gas behaves like an ideal gas (obeys the gas laws), and the con-
servation of energy can be applied. Since the tank often has some insulation and
since the operation of the rocket engines is short, the expansion process will be
close to an adiabatic one. The expansion of the gas in the storage tank causes a
significant drop in gas temperature and an increase in gas density. Furthermore
it assumes that there is no evaporation of the liquid propellant (usually valid for
a propellant with low vapor pressure), the gas is inert, does not dissolve in the
liquid propellant, and there is no sloshing or vortexing.

Let the initial condition in the gas tank be given by subscript 0 and the
instantaneous conditions in the gas tank by subscript g and in the propellant tank
by subscript p. The gas energy after and before propellant expulsion is

mgCvTg+mpCvTp+ppVp=moCyTo (6-5)

The work done by the gas in displacing the propellants is given by PpVp.
Using Eqs. 3-3 to 3-6, the initial storage gas mass nto may be found:

-|- CyPpVp j R “f ppVp — uiqCdTq
m = iPgVo+ppVpk)/{RTo)

( 6 - 6 )

Here c„ is the specific heat at constant volume, R is the gas constant per unit
mass, and k is the specific heat ratio. This equation may be expressed as

Pgm Pp^P j. ^ FpFp / k
Po RTq RTq \l-Pg/po

(6-7)

6.6. TURBOPUMP FEED SYSTEMS AND ENGINE CYCLES 221

The first term in this equation expresses the mass of gas required to empty
a completely filled propellant tank if the gas temperature is maintained at the
initial storage temperature 7 q. The second term expresses the availability of the
storage gas as a function of the pressure ratio through which the gas expands.

Example 6-2. Determine the approximate mass and volume of air needed to pressurize
a tank of 250 kg of 90% hydrogen peroxide for an initial gas tank pressure po of 14 MPa
and a required propellant tank pressure of 3.40 MPa. The density of this propellant is
1388 kg/m^, the gas constant R is 289 J/kg-K, the ambient temperature is 292 K, and for
air ^ = 1 .40.

SOLUTION. Since the vapor pressure of 90% hydrogen peroxide is low, the amount of
liquid propellant that would evaporate is very small and the assumption of no evapora-
tion can apply. The propellant volume is 250/1388 = 0.180 m^. Allow 5% more gas to
compensate for the ambient temperature changes of the liquid propellant and the stored
gas, and for geometric variations of the hardware. From Eq. 6-7 the initial mass of air
mo is

PpVp k

mo = — —

RTa 1 - (Ps/pq)

3.4 X 10® X 0.180 X 1.05 x 1.40
“ 289 X 298 X [1 - (3.4/14)]

= 13.8 kg of compressed air

The volume of the gas tank is from Eq. 3-4:

Vo = nioRTo/po = 13.8 x 289 x 298/(14 x 10®) = 0.0847 m^

The high-pressure gas tank volume is almost half of the propellant tank volume.

6.6. TURBOPUMP FEED SYSTEMS AND ENGINE CYCLES

The principal components of a rocket engine with one type of turbopump system
are shown in the simplified diagram of Fig. 1 -4. Here the propellants are pres-
surized by means of pumps, which in turn are driven by turbines. These turbines
derive their power from the expansion of warm gases.

Figures 10-1, 10-2, and 10-3 show examples of turbopumps and Chapter 10
is devoted exclusively to the turbopump. It is a high precision accurately bal-
anced piece of high shaft speed (rpm) rotating machinery. It has usually two or
one centrifugal pump and a turbine. Its high-speed high-load bearings support the
shaft(s) on which the pump(s) and turbine are mounted. It has shaft seals to pre-
vent leakage of propellants and also to prevent the two propellants from mixing
with each other inside the turbopump. Some turbopumps also have a gear trans-
mission, which allows the pumps or turbine to rotate at a different, usually more

222 LIQUID PROPELLANT ROCKET ENGINE FUNDAMENTALS

efficient, shaft speed. Chapter 10 describes the design of turbopumps, the several
arrangements of the key components, the design of its major components, and
alternate configurations. While pressurized feed systems can be started relatively
quickly, the starting of turbopump feed systems usually takes longer, because it
takes some time for the rotating components (pumps, turbines) to accelerate to
operating shaft speed. Starting is discussed in Section 11.4.

Engines with turbopumps are preferred for booster and sustainer stages of
space launch vehicles, long-range missiles, and in the past also for aircraft
performance augmentation. They are usually lighter than other types for these
high-thrust, long-duration applications. The inert hardware mass of the rocket
engine (without tanks) is essentially independent of duration. Examples can be
seen in Eigs. 6-1 and 6-9 and also in Refs. 6-1, and 6-2. Eor aircraft perfor-
mance augmentation the rocket pump was driven directly by the jet engine, as
in Ref. 6-13. Erom the turbopump feed system options depicted in Chapter 10,
the designer can select the most suitable concept for a particular application.

As stated elsewhere in this book turbopump feed systems are usually preferred
when the engine has a relatively high total impulse, which means high thrust and
long duration. Pressurized feed systems are best for rocket engines with relatively
low total impulse, that is, low thrust and often low cumulative firing durations.

Engine Cycles

Engine cycles apply only to liquid propellant rocket engines with a turbopump
feed system. The three most common engine cycles are shown in Eig. 6-9 and
many of each of these have flown. Reference 6-14 shows variations of these
three cycles and other feasible cycles; some of them have not yet flown and
some have not yet even been built.

An engine cycle for turbopump-fed engines describes the specific propellant
flow paths through the major engine components, the method of providing the hot
gas to one or more turbines, and the method of handling the turbine exhaust gases.
There are open cycles and closed cycles. Open denotes that the working fluid
exhausting from the turbine is discharged overboard, usually after having been
expanded in a separate nozzle of its own as in Eigs. 1-4 and 6-10, or discharged
into the nozzle exit section of the thrust chamber at a point in the expanding
section far downstream of the nozzle throat as shown in Eig. 6-9. In closed
cycles or topping cycles all the working fluid from the turbine is injected into
the engine combustion chamber to make the most efficient use of its remaining
energy. In closed cycles the turbine exhaust gas is fed into the injector of the
thrust chamber and is expanded through the full pressure ratio of the main thrust
chamber nozzle, thus giving a little more performance than the open cycles,
where these exhaust gases expand only through a relatively small pressure ratio.

Table 6-6 shows key parameters for each of the three common cycles and
it describes the differences between them. The schematic diagrams of Eig. 6-9
show each cycle with a separate turbopump for fuel and for oxidier. However,
an arrangement with the fuel and oxdizer pump driven by the same turbine is

6.6. TURBOPUMP FEED SYSTEMS AND ENGINE CYCLES 223

FIGURE 6-9. Simplified diagrams of three engine cycles for liquid propellant rocket
engines. The spirals are a symbol for an axisymmetric cooling jacket where heat is
absorbed.

also feasible and sometimes reduces the hardware mass, volume, and cost. The
“best” cycle has to be selected on the basis of the mission, the suitability of
existing engines, and the criteria established for the particular vehicle. There is
an optimum chamber pressure and an optimum mixture ratio for each application,
engine cycle, or optimization criterion, such as maximum range, lowest cost, or
highest payload.

The gas generator cycle has been the most common. Compared to other engine
cycles, it is relatively simple, pressures are usually lower, generally has a lower
inert mass, and its engine cost is usually lower. However, its performance (spe-
cihc impulse) is lower by a few percent. Its performance is adequate for many
space flight and military missions.

TABLE 6-6. Qualitative Characteristics for Three Different Engine Cycles

c 3

^ I

^ o
*3!

.£

3 o

£ 3

(D ^

Oh (D
?

OJ Oh .

3 o

=2 '3

00 g

in D
ON Cd

-H (D

c« ^

^ e

c§

^ &n
„ - O

a Q
z >
a -3
o

•o

o ^

13 o

o o
cn

HH ^ '

^ (N

M3 ^

»n

.£

-a

f/3

cO o

00 c

(D X

1 3

J-H f/3 . i-H

O ^ T3 O

00 ® o Of

X3 f/3 O ‘X

® S 13 ®

& S

S 3
U

O CO
NO NO
ON ON

O O

c c
3 3

o o
-1^

CJ CJ

o o
D< Di

>, >>

0 o

3 3
3 3

1 ^

&H CH

c/3

^ o

^ I

^ in

a rn

■S) s

o a

o C!h
CO
Vh

t3 O
O

-P N

I !§

^ O
(D cO

Z 3

3 (U

u 0<
Pi

I I

Sh O
QJ GS

:>.

O P-*
00 o

C« >
00

r- ^

•S Pi
o

m E

ON o
^ U ON

<, (D ON
C/5 ^ ^

O 3 ^

"O £

53 -S I

i§ I

.9 O

*ob

3 O

3 3 ?5§

£ 3 ^

0^ c/3
CJ CJ CO

CO • ^

■n C >

CJ 01

c/3 c/3

... cO o

c “.S
M 'o ■£

C 3

W H

■3 X
3 °

Oh CO
O CJ ’

Oh Oh

Oh O 00

M 3 ”0

c/3 >— I
O O Oh
l-i Oh O
Oh o §

S-H O

Oh Oh

£ o

I s

i.s

-C Xi

'oo

“ to

c/3 }_

(U O
Oh

•o o

3 x:
O 00

t-(

00 <n

‘ tfl tfl
Vh U,

3 3

224

Rossiyskiy Naucho Isseldovatelskiy Institut (Reaction Propulsion Research Institute).

6.6. TURBOPUMP FEED SYSTEMS AND ENGINE CYCLES 225

In the gas generator cycle the turbine inlet gas comes from a separate gas
generator. Its propellants can be supplied from two separate propellant tanks or
can be bled off the main propellant feed system tanks. Some early engines used a
separate monopropellant for creating the generator gas. The German V-2 missile
engine used hydrogen peroxide, which was decomposed by a catalyst. Typically,
the turbine exhaust gas is discharged overboard through one or two separate small
low-area-ratio nozzles (at relatively low specific impulse), as shown schematically
in Fig. 1-4 and in the Vulcain engine or RS-68 engine listed in Table 11-2. See
Ref. 6-15. Alternatively, this turbine exhaust can be aspirated into the main flow
through multiple openings in the diverging nozzle section, as shown schematically
in Fig. 6-9 for a gas generator engine cycle. This gas then protects the walls near
the nozzle exit from high temperatures. Both methods can provide a small amount
of additional thrust. The gas generator mixture ratio is usually fuel rich (in some
engines it is oxidizer rich) so that the gas temperatures are low enough (typically
900 to 1350 K) to allow the use of uncooled turbine blades and uncooled nozzle
exit segments. The RS-68 rocket engine, shown in Fig. 6-10, has a simple
gas generator cycle. As can be seen from the data under the figure, with a gas
generator cycle the specific impulse of the thrust chamber by itself is always a
little higher than that of the engine, and the thrust of the thrust chamber is always
slightly lower than that of the engine. The thrust of the RS-68 can be reduced or
throttled to less then 60% of full thrust. See Ref. 6-15.

The expander cycle and the staged combustion cycle are both closed cycles,
and they offer several percent higher performance than the gas generator cycle;
this small improvement makes a substantial difference in payload for flight mis-
sions with high mission velocities. Alternatively they allow a somewhat smaller
flight vehicle. However, the engines are usually more complex, heavier, and more
expensive.

A flow diagram of an expander cycle is shown in Fig. 6-11. The engine
coolant (usually hydrogen fuel) is evaporated, heated, and then fed to low-
pressure-ratio turbines after having passed through the cooling jacket where it
picked up energy. Part of the coolant, perhaps 5 to 15%, bypasses the turbine
(shown in Fig. 6-11) and rejoins the turbine exhaust flow before the entire
coolant flow is injected into the engine combustion chamber where it mixes and
bums with the oxidizer (see Refs. 6-14 and 6-16). The primary advantages of
the expander cycle are good specific impulse, relative engine simplicity, no gas
generator, and relatively low engine mass. In the expander cycle all the propel-
lants are fully burned in the engine combustion chamber and expanded efficiently
in the exhaust nozzle of the thrust chamber.

This cycle is used in the RLIO hydrogen/oxygen rocket engine, and different
versions of this engine have flown successfully in the upper stages of several
space launch vehicles. Data on the RLIO engines are given in Tables 6-8, 8-1,
and 11-2. A modification of this engine, the RLlOB-2 with an extendible nozzle
skirt, can be seen in Fig. 8-17. The RLlOB-2 flow diagram in Fig. 6-11 shows

226 LIQUID PROPELLANT ROCKET ENGINE EUNDAMENTALS

Roll control
nozzle using
turbine
exhaust

Gimbal actuator

Oxidizer valve

Regenerative cooled thrust
chamber with nozzle throat

Ablative lined
bell-shaped nozzle
extension

High pressure
oxygen line

Oxygen

turbopump

Oxygen tank

pressurization

heat

exchanger

Turbine

exhaust

nozzle

Main fuel
valve

Eour leg structural
support

Fuel turbopump

Parameter

Thrust chamber

Engine

Specific impulse at sea level (max.), sec

363

357

Specific impulse in vacuum (max.), sec

415

409

Thrust, at sea level, Ibf

640,700

656,000

Thrust in vacuum Ibf

732,400

751,000

Mixture ratio

6.74

6.0

FIGURE 6-10. Simplified view of the RS-68 rocket engine with a gas generator cycle.
For engine data see Table 10-3. (Courtesy of Pratt & Whitney, Rocketdyne.)

its expander cycle. The turbine drives a single-stage liquid oxygen pump (through
a gear case) and a two-stage liquid hydrogen pump. The cooling down of the
hardware to cryogenic temperatures is accomplished by flowing (prior to engine
start) cold propellant through cool-down valves. The pipes for discharging the
cooling propellants overboard are not shown here but can be seen in Fig. 8-17.

6.6. TURBOPUMP FEED SYSTEMS AND ENGINE CYCLES 227

Oxidizer flow
control valve

Oxygen pump

Liquid

oxygen

Propellant
shutoff valves

Liquid

hydrogen

Fuel pump

Fuel cooldown and
pressure relief valve

FIGURE 6-11. Simplified schematic flow diagram of the RLlOB-2 upper-stage rocket
engine is an example of an expander engine cycle. For data see Tables 6-8 and 8-1.
(Courtesy of Pratt & Whitney Rocketdyne.)

Thrust is regulated by controlling the flow of hydrogen gas to the turbine, using
a bypass to maintain constant chamber pressure. Helium is used as a means of
power boost by actuating several of the larger valves through solenoid-operated
pilot valves.

In the staged combustion cycle, the coolant flow path through the cooling
jacket is the same as that of the expander cycle. Here a high-pressure preburner
(a high-pressure gas generator) burns all the fuel with part of the oxidizer to
provide high-energy gas to the turbines. The total turbine exhaust gas flow is
injected into the main combustion chamber where it burns with the remaining
oxidizer. This cycle lends itself to high-chamber-pressure operation, which allows
a small thrust chamber size. The extra pressure drop in the preburner and turbines
causes the pump discharge pressures of both the fuel and the oxidizer to be higher
than with open cycles, requiring heavier and more complex pumps, turbines, and
piping. The turbine flow is relatively high and the turbine pressure drop is low,
when compared to other cycles. The staged combustion cycle gives high specific
impulse, but it is more complex and heavy. A variation of the staged combustion
cycle is used in the Space Shuttle Main Engine, as shown in Figs. 6-1 and 6-12.
This engine actually uses two separate preburner chambers, each mounted directly
on a separate main turbopump. In addition, there are two more turbopumps for
providing a boost pressure to the main pumps, but their turbines are not driven by
combustion gases; instead, high-pressure liquid oxygen drives one booster pump
and evaporated hydrogen drives the other. The injector of this reusable liquid
propellant high-pressure engine is shown in Fig. 9-6 and performance data are

228 LIQUID PROPELLANT ROCKET ENGINE FUNDAMENTALS

Low pressure

Hydrogen t fuel (booster) ^ Low pressure t Oxygen

^riy^l S ? iiv\/nQn riQo ' fc- ~ . ■ ■

Hydrogen gas bleed_^
for pressurizing | '
the fuel tank ^

One of two
oxygen control
valves of the
preburners

Thrust chamber
gas exhaust

FIGURE 6-12. Flow diagram illustrating the staged combustion cycle of the Space
Shuttle Main Engine (SSME) using liquid oxygen and liquid hydrogen fuel. (Courtesy of
Pratt & Whitney Rocketdyne and NASA.)

given in Tables 10-1 and 11-2. While the Space Shuttle Main Engine (burning
hydrogen with oxygen) has fuel-rich preburners, oxidizer-rich preburners are used
in the Energomash RD-170 and RD-120 engines (kerosene/oxygen) and other
Russian staged combustion rocket engines. See Table 11-3. Another example
of a staged combustion cycle is the Russian engine RD-253; all of the nitrogen
tetroxide oxidizer and some of the unsymmetrical dimethyl hydrazine fuel are
burned in the preburner, and the remaining fuel is injeeted directly into the main
combustion chamber.

In an engine with a gas generator cycle the thrust of the thrust chamber is
always slightly less than the Thrust of the engine, because the exhaust from the
turbine provides a little extra thrust. Also the thrust chamber specihe impulse
is a little higher than the rocket engine specific impulse, because the low-flow
gas exhausted from the turbine has a very low specific impulse. Eor closed
cycles, namely the expander cycle and the staged combustion cycle, the values
of the thrust and the specific impulse are the same for the engine and its thrust
chamber(s).

6.7. ROCKET ENGINES EOR MANEUVERING, ORBIT ADJUSTMENTS, OR ATTITUDE CONTROL 229

6.7. ROCKET ENGINES FOR MANEUVERING, ORBIT ADJUSTMENTS,
OR ATTITUDE CONTROL

These engines have usually a set of small thrusters that are installed at various
places in a vehicle and a common pressurized feed system, similar to Figs. 1-3,
4-13, or 6-13. They are called reaction control systems or auxiliary rocket
propulsion systems as contrasted to higher-thrust primary or boost propulsion
systems. Most have low thrust, use storable liquid propellants, require accurate
repeatable pulsing, a long life in space, and/or a long-term storage with loaded
propellants in flight tanks. Typical thrust levels of a small thruster are between 0. 1
and 1000 Ibf. Figure 4-13 shows that it requires 12 thrusters for the application of
pure torques about three vehicle axes. If a three-degree-of-rotation freedom is not
a requirement, or if torques can be combined with some translation maneuvers,
fewer thrusters will be needed. These auxiliary rocket engines are commonly
used in spacecraft or missiles for the accurate control of flight trajectories, orbit
adjustments , or attitude control of the vehicle. References 6-1 and 6-17 give
information on several of these. Figure 6-13 shows a simplifled flow diagram
for a postboost control rocket engine, with one larger rocket thrust chamber for
changing the velocity vector and eight small thrusters for attitude control.

Section 4.5 describes various space trajectory correction maneuvers and satel-
lite station-keeping maneuvers that are typically performed by these small aux-
iliary liquid propellant rocket engines with multiple thrusters. Table 6-7 lists
typical applications for rocket engines with small thrusters.

Attitude control can be provided both while a primary propulsion system (of
a vehicle or of a stage) is operating and while its small thruster rocket system
operates by itself. For instance, this is done to point satellite’s telescope into

TABLE 6-7. Typical Applications for Small Thrusters

Flight path (or orbit) corrections or changes
Minor flight velocity adjustments
Station keeping (correcting for deviations from orbit)

Orbit injection for small satellites
Deorbit maneuver for small satellites
Divert maneuvers of terminal interceptors
Attitude control for:

Satellites, stages of space launch vehicles, space stations, missiles
Roll control for a single gimbaled rocket engine
Pointing/orienting antennas, solar cells, mirrors, telescopes, etc.

Correct the misalignment of principal, larger thrust chamber
Velocity tuning of war heads (postboost control system) for accurate targeting
Settling of liquid propellants prior to gravity-free start of main upper stage engine
Flywheel desaturation

Source: Mostly from Ref. 6-17.

230 LIQUID PROPELLANT ROCKET ENGINE EUNDAMENTALS

a specific orientation or to rotate a spacecraft’s main thrust chamber into the
desired direction for a vehicle turning maneuver.

A common method for achieving accurate velocity corrections or precise angu-
lar positions is to operate or fire some of the thrusters in a pulsing mode (e.g.,
fire repeatedly for 0.020 sec, each time followed by a pause of perhaps 0.020 to
0.100 sec). The guidance system determines the maneuver to be undertaken and
the vehicle control system sends command signals to specific thrusters for the
number of pulses needed to accomplish this maneuver. Small liquid propellant
engine systems are uniquely capable of these pulsing operations. Some small
thrusters have been tested for more than 300,000 pulses. For very short pulse
durations the specific impulse is degraded by 5 to 25%, because the performance
during the thrust buildup and thrust decay period (at lower chamber pressure) is
inferior to operating only at the rated chamber pressure and the transient time
becomes a major portion of the total pulse time.

Ballistic missile defense vehicles usually have highly maneuverable upper
stages. These require substantial side forces also called divert forces (200 to
6000 N) during the final closing maneuvers just prior to reaching the target.
In concept the system is similar to that of Fig. 6-13, except that the larger
thrust chamber would be at right angles to the vehicle axis. A similar system for
terminal maneuvers, but using solid propellants, is shown in Figs. 12-27 and
12-28.

The Space Shuttle performs its reaction control with 38 different thrusters,
as shown schematically in Figs. 1-14 and 6-8; this includes several duplicate
(spare or redundant) thrusters. Selected thrusters are used for different maneuvers,
such as space orbit corrections, station keeping, or positioning the Space Shuttle
for reentry or visual observations. These small restartable rocket engines are also
used for space rendezvous or docking maneuvers , where one spacecraft slowly
approaches another and locks itself to the other, without causing excessive impact
forces during this docking manuever. This docking operation requires rotational
and translational maneuvers from a series of rocket engines.

The application of pure torque to spacecraft can be divided into two classes,
mass expulsion types (rockets) and nonmass expulsion types. Nonmass expulsion
types include momentum storage, gravity gradient, solar radiation, and magnetic
systems. Some space satellites are equipped with both the mass and nonmass
expulsion types. Reaction wheels or flywheels, a momentum storage device, are
particularly well suited to obtaining vehicle angular position control with high
accuracies of less than 0.01° deviation and low vehicle angular rates of less
than 10“^ degrees/sec with relatively little expenditure of energy. The vehicle
angular momentum is changed by accelerating (or decelerating) the wheel. Of
course, when the wheel speed reaches the maximum (or minimum) permissible,
no further electrical motor torquing is possible; the wheel must be decelerated (or
accelerated) to have its momentum removed (or augmented), a function usually
accomplished through the simultaneous use of two small attitude control rocket
thrusters, which apply a torque to the vehicle in the opposite direction. This has
been called desaturation of the fly wheel.

6.7. ROCKET ENGINES EOR MANEUVERING, ORBIT ADJUSTMENTS, OR ATTITUDE CONTROL 231

FIGURE 6-13. Schematic flow diagram of the helium-pressurized, bipropellant rocket
engine system of the fourth stage of the Peacekeeper ballistic missile, which provides
the terminal velocity (in direction and magnitude) to each of several warheads. It has
one larger gimballed thrust chamber for trajectory translation maneuvers and eight small
thrusters (with scarfed nozzles) for attitude control in pitch, yaw, and roll. For clarity the
tanks and feed system are shown outside the vehicle skin; really they are located within
the vehicle. (Courtesy of USAF.)

232 LIQUID PROPELLANT ROCKET ENGINE EUNDAMENTALS

The propellants for auxiliary rockets fall into three categories: cold gas jets
(also called inert gas jets), warm or heated gas jets, and chemical combustion
rockets, such as bipropellant liquid propellant rockets. The specific impulse is
typically 50 to 120 sec for various cold gas systems and 105 to 250 sec for
warm gas systems. Warm gas systems can use inert gas with an electric heater
or a monopropellant which is catalytically and/or thermally decomposed. Bipro-
pellant attitude control thrust chambers allow an /j of 220 to 325 sec and have
varied from 5 to 4000 N thrust; the highest thrusts apply to large spacecraft. All
basically use pressurized feed systems with multiple thrusters or thrust cham-
bers equipped with fast-acting, positive-closing precision valves. Many systems
use small, uncooled, metal-constructed thrusters with supersonic exhaust nozzles,
and they are strategically located on the periphery of the spacecraft pointing in
different directions. Gas jets are used typically for low thrust (up to 10 N) and
low total impulse (up to 4000 N-sec). They have been used on smaller satellites
and often only for roll control. See Ref. 6-17 and Section 7.6.

Small liquid monopropellant and liquid bipropellant rocket units are common
in auxiliary rocket systems for thrust levels typically above 2 N and total impulse
values above 3000 N-sec. Hydrazine is the most common monopropellant used
in auxiliary control rockets; nitrogen tetroxide and monomethylhydrazine is a
common bipropellant combination. The next chapter contains data on all three
categories of these propellants, and Chapter 8 shows diagrams of small thrusters.

Combination systems are also in use. Here a bipropellant with a relatively high
value of Is, such as N2O4 and N2H4, is used in the larger thrusters, which consume
most of the propellant; then several simple monopropellant thrusters (with a
lower Is), used for attitude control pulsing, usually consume a relatively small
fraction of the total fuel. Another combination system is to employ bipropellant
or monopropellant thrusters for adding a velocity increment to a flight vehicle
or to bleed or pulse some of the pressurizing gas, such as helium, through small
nozzles controlled by electromagnetic valves to provide roll control. The specific
mission requirements need to be analyzed to determine which type or combination
is most advantageous for a particular application.

Special thruster designs exist that can be used in a bipropellant mode at higher
thrust and also in a monopropellant mode for lower thrust. This can offer an
advantage in some spacecraft applications. An example is the Northrop-Grumman
secondary combustion augmented thruster (SCAT), which uses hydrazine and
nitrogen tetroxide, is restartable, vaporizes the propellants prior to injection and
therefore has very efficient combustion (over 99%), can operate over a wide range
of mixture ratios, and can be throttled from 5 to 15 Ibf thrust. From Ref. 6-12.

6.8. ENGINE FAMILIES 233

6.8. ENGINE FAMILIES

An engine family is a series of related rocket engines, which have evolved over a
period of many years. They come from the same rocket engine organization and
each engine has been tailored to a specific application. The engines in the family
have a strong resemblance to each other, use the same engine concept, usually
the same propellants, and some identical or somewhat modified components of
the same type. When an existing proven liquid propellant rocket engine can be
modified and/or up-rated to fit a new application, the newer modified engine can
use a lot of the proven hardware, test data, and software of the original earlier
engines.

An example is the RL 10 rocket engine family of upper-stage engines. It was
developed by Pratt & Whitney Rocketdyne over a period of about 46 years and
is shown in Table 6-8. Data is from Refs. 6-12 and 6-16. Each engine is for
a specific application and is a modification and/or uprating of an earlier model.
The principal changes from one engine model to the next included increases in
thrust, increases in performance (somewhat higher specific impulse) by using
higher chamber pressure, improved injector designs, and increases in nozzle exit
area ratios. All use LOX/LH 2 propellants, the same basic engine concept with an
expander engine cycle, the same tubular cooling jacket approach for the chamber
and nozzle throat region, the same generic geared turbopump arrangement, often
the same or similar valves, and power level control by a bypass of hydrogen gas
around the turbine. In the turbopump the fuel pump and turbine are on the same
high-speed shaft and the LOX pump is driven efficiently through a gear train at
a lower speed. All engines are gimbal mounted (most at 4° maximum deflection)
and most have space restart capability. Figure 8-17 shows the extendable nozzle
of the RL lOB-2. In the RL 10A-3-3A thrust chamber a high-conductivity silver
ring was brazed with silver into the nozzle throat, thus enabling a higher chamber
pressure. The cooling jacket tubes were brazed with silver and were compatible.
Figure 6-11 shows a flow sheet of an RL 10 expander engine cycle.

It is interesting to see how thrust or specific impulse changed with time and
with the model. Incidentally the specific impulse listed in Table 6-8 for the
RL lOB-2 is the highest of any flying liquid propellant rocket engine in the world
and the extendable nozzle exit segment of the RL lOA-4 was the first for liquid
propellant rocket engines.

When compared to a brand new engine, the principal benefits of adopting
a modified engine, which is based on an earlier family of proven engines, are
savings in costs (costs of design, development, less fabrication, less testing, quali-
fication, and operation), attaining a high engine reliability more quickly, and often
also a shorter schedule. The heritage of earlier proven similar engines allows
the use of old engine or component data, having trained experienced personnel,
proven subcontractors, a higher confidenee level of reliability, and often, but not

Engine Qualification pi Thrust Mixture Weight 4 (vac) Nozzle No. Engines

Model Year Vehicle psia Ibf Ratio Ibf sec Area Ratio per Stage Comments

C £

0 W

H s

> -c .

cj m

o § 3

^ — ,«5
0) ^

CQ C/!)

a « o
^ > o
C i-

^ I “

1 s I

IC (J x:

_£ c
o %
^ o

D 7D

"o
5S O
P. ^
1)

« g

ffl

U j,

Q, Sj

Q. c/n

— O
"O
a O
a. o

u ■

>. •

I ~a
' -p

■s

Cl *-

s e.

3 u.
Cl s
O

"P e

E-

.2 N

o OJ • —

g- 00

c ^ 52
£ > -g
3 rp
Q. C/5 C

U S
5 D

t:
c

u -a S

D H
■d

?3 lT

La — /-s

C W)
1> a
c >»

'§ :> '

S "O '

1) "O

^ C

o ^
.5 X)

-5 Z

> .£
C,

00 >

> o

^ ® 3 ^

•a -c 2
S o

§ ° s-

■ 3 II 8
' s S 6 f
S .S 2 S

o ^
^ (50

■K s=

■£ -c

(• i

c/5

—

c/5

C/5

234

RL 10A-3-3B 1986 Shuttle Centaur 425 15,000 6.0;1 310 436 61.0;1 2 RL 10A-3-3A modified to handle long

space stay time for AF version of
Shuttle/Centaur. Mixture ratio
increased to 6.0 and cooldown
components changed. Never flew.

c/3

>» c

0 ) O "O

-D

^ B

TD X
O

9- S

C3 ^

O —H
£■ ^
® J -

o <

•o >

U - r -

> cs

-S OJ 5=

w ^ .2

Q. ^ c«

^ e C
^

^ ^ X
tn 0 )

o .O-
•o U
cn *2
^ 1j
O Q

rr. W

^ c
t ? ^

3 O
J= 'C

^ .3

■s a

00 C/3 c/3

3 G C

■ 5--S s

S g

55 cc o

0 « ;S "O

. 2P S

-C o
•o

•o

3 o c
c ^
§ cn g

N 3
O «

c ®

0) (U

1 1
3 TD
-C C

>< C >

O ^

d ^

m IT )
(N O"
(N (N

3 3

C G

u u

c/3 C

c/5

—

< P

■3" 0^

< <
o o

235

236 LIQUID PROPELLANT ROCKET ENGINE FUNDAMENTALS

always, the use of the same debugged materials, fabrication, and test facilities.
An intangible benefit is that the vehicle developer or prime contractor, who plans
to use one of these engines, will have more confidence in the engine. A brand
new engine may have slightly better performance and/or a somewhat lower inert
engine mass and some other improvements, but it will usually be more costly,
take longer to develop, and take more time to reach an equivalent high level of
reliability.

6.9. VALVES AND PIPELINES

Valves control the flows of liquids and gases and pipes conduct these fluids to
the intended components. There are no rocket engines without them. There are
many different types of valves. All have to be reliable, lightweight, leakproof,
and must withstand intensive vibrations and very loud noises. Table 6-9 gives
several key classification categories for rocket engine valves. Any one engine
will use only some of the valves listed here.

The art of designing and making valves is based, to a large extent, on expe-
rience. A single chapter cannot do justice to it by describing valve design and
operation. References 6-1 and 6-2 decribe the design of specific valves, lines,
and joints. Often the design details, such as clearance, seat materials, or opening
time delay present development difficulties. With many of these valves, any sig-
nificant leakage or valve failure can cause a failure of the engine itself. All valves
are tested for two qualities prior to installation; they are tested for leaks — through
the seat and also through the glands — and for functional soundness or perfor-
mance.

The propellant valves in high-thrust units handle relatively large flows at high
service pressures. Therefore, the forces necessary to actuate the valves are large.
Hydraulic or pneumatic pressure, controlled by pilot valves, operates the larger
valves; these pilot valves are in turn actuated by a solenoid or a mechanical
linkage. Essentially this is a means of power boost.

Two valves commonly used in pressurized feed systems are isolation valves
(when shut, they isolate or shut off a portion of the propulsion system) and latch
valves; they require power for brief periods during movements, such as to open
or shut, but need no power when latched or fastened into either an open or a
closed position.

A very simple and very light valve is a burst diaphragm . It is essentially a
circular disk of material that blocks a pipeline and is designed so that it will
fail and burst at a predetermined pressure differential. Burst diaphragms are
positive seals and prevent leakage, but they can be used only once. The German
Wassetfall anti-aircraft missile, an early application, used four burst disks; two
were in high-pressure air lines and two were in the propellant lines.

6.9. VALVES AND PIPELINES 237

TABLE 6-9. Classification of Valves Used in Liquid Propellant Rocket Engines"

1 . Fluid : fuel; oxidizer; cold or heated pressurized gas; hot turbine gas

2. Application or Use: main propellant control; thrust chamber valve (dual or single);
bleed; vent; drain; fill; by-pass; preliminary stage flow; pilot valve; safety valve; over-
board dump; regulator; gas generator or preburner control; sequence control; prevent
back flow; isolation of propellant or, latch valve

3. Mode of Actuation: automatically operated (by solenoid, pilot valve, trip mechanism,
pyrotechnic, etc.); manually operated; pressure-operated by air, gas, propellant, or
hydraulic fluid (e.g., check valve, tank vent valve, pressure regulator, relief valve), with
or without position feedback, rotary or linear actuator

4. The Flow magnitude determines the size of the valve

5. Duty cycle: single operation or multiple operation during the same flight, short duration
pulse operation; reusable for other flights; long or short life

6. Valve Type: normally open; normally closed; normally partly open; two-way; three-
way, with/without valve position feedback; ball valve, gate valve, butterfly type, spring
loaded, low pressure drops

7. Temperature and pressure allow classification by high, low, or cryogenic temperature
fluids, or high or low pressure or vacuum capability

8. Accessible or not accessible to inspection, servicing, or replacement of valve or its seal

"This list is not comprehensive or complete.

Figure 6-14 shows a large main liquid oxygen valve. It is normally closed,
rotary actuated, cryogenic, high pressure, high flow, reusable ball valve with a
low-pressure loss in the open position, allowing continuous throtting, a controlled
rate of opening through a crank and hydraulic piston (not shown), with a position
feedback and anti-icing controls.

Pressure regulators are special valves that are used frequently to regulate
gas pressures. Usually the discharge pressure is regulated to a predetermined
standard pressure value by continuously throttling the flow, using a piston, flexible
diaphragm, or electromagnet as the actuating mechanism. Regulators can be seen
in the flow sheets of Figs. 1-3 and 6-13.

The various fluids in a rocket engine are usually transported by pipes or
lines, usually made of metal and joined by fittings or welds. Their design must
provide for thermal expansion and provide support to minimize vibration effects.
For gimballed thrust chambers it is necessary to provide flexibility in the piping
to allow the thrust axis to be rotated through a small angle, typically ±3 to
±10°. This flexibility is provided by flexible pipe joints and/or by allowing
pipes to deflect slightly when using two or more right-angle turns in the lines.
The high-pressure propellant feed lines of the SSME have both flexible joints
and right-angle bends, as shown in Figs. 6-1 and 6-15. This joint has flexible
bellows as a seal and a universal joint-type mechanical linkage with two sets of
bearings for carrying the separating loads imposed by the high pressure.

238 LIQUID PROPELLANT ROCKET ENGINE FUNDAMENTALS

Valve

outlet

Shaft seal

bearing

Intergral

ball/shatt/cams

Bearing

Inlet

seal

Cam follower
pin and bearing

Section A-A

Hydraulic

actuator

housing

Thermal.

Insulator

Valve

housing

Closed

Seal

liftoff

Open

FIGURE 6-14. The SSME main oxidizer valve is a low-pressure drop ball valve rep-
resentative of high-pessure large valves used in rocket engines. The ball and its integral
shaft rotate in two bearings. The seal is a machined plastic ring spring-loaded by a bel-
lows against the inlet side of the ball. Two cams on the shaft lift the seal a short distance
off the ball within the first few degrees of ball rotation. The ball is rotated by a preci-
sion hydraulic actuator (not shown) through an insulating coupling. (Courtesy of Pratt &
Whitney Rocketdyne.)

Sudden closing of valves can cause water hammer in the pipelines, leading
to unexpected pressure rises that can be destructive to propellant system compo-
nents. An analysis of this water hammer phenomenon will allow determination of
the approximate maximum pressure (Ref. 6-18). The friction of the pipe and the
branching of pipelines reduce this maximum pressure. Water hammer can also
occur when admitting the initial flow of high-pressure propellant into evacuated
pipes. The pipes are under vacuum to remove air and prevent the forming of gas
bubbles in the propellant flow, which can cause combustion problems.

Many liquid rocket engines have filters in their lines. This is necessary to
prevent dirt, particles, or debris, such as small pieces from burst diaphragms, from
entering precision valves or regulators (where debris can cause a malfunction)
or from plugging small injection holes, which could cause hot streaks in the
combustion gases, in turn causing a thrust chamber failure.

Occasionally, a convergent-divergent venturi section, with a sonic velocity
at its throat, is placed into one or both of the liquid propellant lines. It has
also been called a cavitating venturi, when the local throat pressure goes below

6.10. ENGINE SUPPORT STRUCTURE 239

Bearing Bellows seal

Sleeve

FIGURE 6-15. Flexible high-pressure joint with external gimbal rings for a high-
pressure hot turbine exhaust gas. (Courtesy of Pratt & Whitney Rocketdyne)

the vapor pressure. The merits are that it maintains constant flow and prevents
pressure disturbances from traveling upstream. This can prevent the propagating
of chamber pressure oscillations or coupling with thrust chamber combustion
instabilities. The venturi section can also help in minimizing some water hammer
effects in a system with multiple banks of thrust chambers.

6.10. ENGINE SUPPORT STRUCTURE

Most of the larger rocket engines have their own mounting structure or support
structure. On it the major components are mounted. It also transmits the thrust
force to the vehicle. Welded tube structures or metal plate/sheet metal assemblies
have been used. In some large engines the thrust chamber is used as a structure
and the turbopump, control boxes, or gimbal actuators are attached to it.

In addition to the thrust load, an engine structure has to withstand forces
imposed by vehicle maneuvers (in some cases a side acceleration of 10 go),
vibration forces, actuator forces for thrust vector control motions, and loads from
transportation over rough roads.

In low-thrust engines with multiple thrusters there often is no separate engine
mounting structure; the major components are in different locations of the vehicle,
connected by tubing, wiring, or piping, and each is usually mounted directly to
the vehicle or spacecraft structure.

240 LIQUID PROPELLANT ROCKET ENGINE EUNDAMENTALS

PROBLEMS

1 . In an engine with a gas generator engine cycle, the turbopump has to do more work in
the pumps, if the thrust chamber operating pressure is raised. This of course requires
an increase in turbine gas flow which, when exhausted, adds little to the engine specific
impulse. If the chamber pressure is raised too much, the decrease in performance due
to an excessive portion of the total propellant flow being sent through the turbine and
the increased mass of the turbopump will outweigh the gain in specific impulse that
can be attained by increased chamber pressure and also by increased thrust chamber
nozzle exit area. Outline in detail a method for determining the optimum chamber
pressure where the sea-level performance will be a maximum for a rocket engine that
operates in principle like the one shown in Fig. 1 -4.

2 . The engine performance data for a turbopump rocket engine system are as follows:

Propellants

Engine system specific impulse (steady state)

Engine system mixture ratio

Rated engine system thrust

Oxidizer vapor flow to pressurize oxidizer

tank

Propellant flow through turbine at rated thrust
Gas generator mixture ratio
Specific impulse of turbine exhaust

Liquid oxygen/kerosene
272 sec
2.52

40,000 N

0.003% of total oxidizer flow

2.1% of total propellant flow
0.23
85 sec

Determine performance of the thrust chamber L,r,F (see Section 11.2).

3 . For a pulsing rocket engine, assume a simplified parabolic pressure rise of 0.005 sec,
a steady-state short period of full chamber pressure, and a parabolic decay of 0.007
sec approximately as shown in the sketch. Plot curves of the following ratios as a
function of operating time t from t = 0.013 to t = 0.200 sec: (a) average pressure
to ideal steady-state pressure (with zero rise or decay time); (b) average L to ideal
steady-state L; (c) average F to ideal steady-state F .

Time

PROBLEMS 241

4 . For a total impulse of 100 Ibf-sec compare the volume and approximate system weights
of a pulsed propulsion system using different gaseous propellants, each with a single
spherical gas storage tank (at 3500 psi and 0°C). A package of small thrust nozzles with
piping, valves, and controls is provided which weighs 3.2 lb. The gaseous propellants
are hydrogen, nitrogen, or argon (see Table 7-3).

5. Compare several systems for a potential roll control application which requires four
thrusters of 1 Ibf each to operate for a cumulative duration of 2 min each over a period
of several days, which allows a constant gas temperature. Include the following:

Pressurized helium 70°F temperature

Pressurized nitrogen 70°F Ambient temperature

Pressurized krypton 70°F Ambient temperature

Pressurized helium at 300°F (electrically heated)

The pressurized gas is stored at 5000 psi in a single spherical fiber-reinforced plastic
tank; use a tensile strength of 150,000 psi and a density of 0.050 Ibm/in.^ with a
0.012-in. -thick aluminum inner liner as a seal against leaks. Neglect the gas volume
in the pipes, valves, and thrusters, but assume the total hardware mass of these to be
about 1.3 Ibm. Use Table 7-3. Make estimates of the tank volume and total system
weight. Discuss the relative merits of these systems.

6. A sealed propellant tank contains hydrazine. It is stored for long periods of time,
and therefore the propellant and the tank will reach thermal equilibrium with the
environment. At an ambient temperature of 20°C and an internal pressure of 1.2 atm
the liquid occupies 87% of the tank volume and the helium pressurizing gas occupies
13%. Assume no evaporation of the propellant, no dissolving of the gas in the liquid,
and no movement of the tank. Use the hydrazine properties from Figs. 7-1 and 7-2
and Table 7-1. What will be the approximate volume percentages and the gas pressure
at the extreme storage temperatures of 4 and 40° C?

7. A liquid hydrogen/liquid oxygen thrust chamber has a constant bipropellant flow rate
of 347 kg/sec at a mixture ratio of 6.0. It operates at full thrust for exactly 2 min. The
propellant in the tanks are initially at the propellant boiling points and are assumed to
be of uniform initial temperature at start. Use data from Table 7-1 for the propellant
specific gravities. Assuming no losses, find the masses of (a) fuel and (b) of oxidizer
used to produce the thrust for the nominal duration, (c) What was the volume of
the liquid hydrogen actually used? (d) Assuming 4.0% extra fuel mass (for unusable
propellant residual, evaporation, hardware cooling, or venting just prior to start, or
propellant consumed inefficiently during startup and shutdown) and a 10% ullage
volume (the void space above the liquid in the tank), what will be the volume of the
fuel tank? Assume other losses can be neglected.

Answers: (a) 5941 kg, (b) 35,691 kg, (c) 83.83 m^ (d) 95.6 m^.

8. Prepare dimensioned rough sketches of the four propellant tanks needed for operating
a single gimbal-mounted RD 253 engine (Table 11-3) for 80 sec at full thrust and

242 LIQUID PROPELLANT ROCKET ENGINE FUNDAMENTALS

an auxiliary rocket system with a separate pressurized feed system using the same
propellants, with two gimbal-mounted small thrust chambers, each of 150 kg thrust, a
duty cycle of 12% (fires only 12% of the time), but for a total flight time of 1.00 hr.
For propellant properties Table 7-1. Describe any assumptions that were made with
the propellant budget, the engines, or the vehicle design, as they affect the amount of
propellant.

9. Table 10-5 shows that the RD 120 rocket engine can operate at thrusts as low as 85%
of full thrust and with a mixture ratio variation of ±10.0%. Assume a 1.0% unavailable
residual propellant. The allowance for operational factors, loading uncertainties, off-
nominal rocket performance, and a contingency is 1.27% for the fuel and 1.15% for
the oxidizer.

(a) In a particular flight the average main thrust was 98.0% of nominal and the mixture
ratio was off by ±2.00% (oxidizer rich). What percent of the total fuel and oxidizer
loaded into the vehicle will remain unused at thrust termination?

(b) If we want to run at a fuel-rich mixture in the last 20% of the flight duration (in
order to use up all the intended flight propellant), what would the mixture ratio
have to be for this last period?

(c) In the worst possible scenario with maximum throttling and extreme mixture ratio
excursion (±3.00%, but operating for the nominal duration), what is the largest
possible amount of unused oxidizer or unused fuel in the tanks?

SYMBOLS

effective exhaust velocity, m/sec (ft/sec)
specific heat at constant volume, J/kg-K (Btu/lbm-°R)
thrust coefficient
thrust force, N (Ibf)

acceleration of gravity at sea level, 9.8066 m/sec^

specific impulse, sec

specific heat ratio

propellant mass, kg (Ibm)

mass flow rate, kg/sec (Ib/sec)

pressure, N/m^ (psi)

pressure difference, N/m^ (psi)

mixture ratio (oxidizer to fuel mass flow rates)

gas constant per unit mass, J/kg-K (ft-lbf/lbm-°R)

absolute temperature, K

volume flow rate, m^/sec (ft^/sec)

total propellant weight, N (Ibf)

weight flow rate, N/sec (Ibf/sec)

REFERENCES 243

Subscripts

fuel

initial condition or stagnation condition

gas tank

oxidizer

propellant tank or power cutoff

REFERENCES

6-1. D. K. Huzel and D. H. Huang, Design of Liquid Propellant Rocket Engines, rev.
ed., AIAA, Reston, VA, 1992.

6-2. G. G. Gakhun, V. I. Baulin, et al. Construction and Design of Liquid Propel-
lant Rocket Engines (in Russian), Konstruksiya i Proyektirovaniye Zhidkostniyk
Raketnykh Dvigateley , Mashinostroyeniye, Moscow, 1989.

6-3. J. I. Hochsten, H.-C. Ji, and J. Ayelott, “Prediction of Self-Pressurization Rate of
Cryogenic Propellant Tankage,” Journal of Propulsion and Power, Vol. 6, No. 1,
January-February 1990, pp. 11-17.

6-4. B. Morton, M. Elgersma, and R. Playter, “Analysis of Booster Vehicle Slosh Sta-
bility During Ascent to Orbit,” AIAA Paper 90-1876, July 1990.

6-5. J. J. Pocha, “Propellant Slosh in Spacecraft and How to Live with It,” Aerospace
Dynamics, Vol. 20, Autumn 1986, pp. 26-31.

6-6. G. P. Purohit and L. D. Loudenback, “Application of Etched Disk Stacks in Surface
Tension Propellant Management Devices,” Journal of Propulsion and Power, Vol.
7, No. 1, January-February 1991, pp. 22-30.

6-7. J. R. Rollins, R. K. Grove, and D. R. Walling, Jr. “Design and Qualification of
a Surface Tension Propellant Tank for an Advanced Spacecraft,” AIAA Paper
88-2848, 24th Joint Propulsion Conference, 1988.

6-8. Design Guide for Pressurized Gas Systems, Vols. I and II, prepared by JIT Research
Institute, NASA Contract NAS7-388, March 1966.

6-9. H. C. Hearn, “Evaluation of Bipropellant Pressurization Concepts for Spacecraft,”
Journal of Spacecraft and Rockets, Vol. 19, July 1982, pp. 320-325.

6-10. H. C. Hearn, “Design and Development of a Large Bipropellant Blow-
down Propulsion System,” Journal of Propulsion and Power, Vol. 11, No. 5,
September-October 1995.

6-11. National Space Transportation System Reference, Vol. 1, National Aeronautics and
Space Administration, Washington, DC, June 1988 (description of Space Shuttle
system and operation).

6-12. Personal communications from Jim Morehart, The Aerospace Corp., Vince Whee-
lock, Pat Mills, and Randy Parsley of Pratt & Whitney Rocketdyne, and Gordon
Dressier, Northrop Grumman Corp.

244 LIQUID PROPELLANT ROCKET ENGINE FUNDAMENTALS

6-13. H. Grosdemange and G. Schaeffer, “The SEPR 844 Reuseable Liquid Rocket
Engine for Mirage Combat Aircraft,” AIAA Paper 90-1835, July 1990.

6-14. D. Manski, C. Goertz, H. D. Sassnick, J. R. Hulka, B. D. Goracke, and D. J. H.
Levack, “Cycles for Earth to Orbit Propulsion,” Journal of Propulsion and Power,
AIAA, Vol. 14, No. 5, September- October 1998.

6-15. D. Conley, N. Y. Lee, P. L. Portanova and B. K. Wood, “Evolved Expendable
Launch Vehicle System: RS-68 Main Engine Development,” Paper 1AC-02.S.1.01,
53rd International Astronautical Congress, 10-19 Oct. 2002, Houston, Texas.

6-16. J. R. Brown, “Expander Cycle Engines for Shuttle Cryogenic Upper Stages,”
AIAA Paper 83-1311, 1983.

6-17. G. P. Sutton, “History of Small Liquid Propellant Thrusters,” presented at 52nd
JANNAF Propulsion Meeting, 12 May 2004, Las Vegas, Nevada; published by
Chemical Propulsion Information Analysis Center, Columbia, MD.

6-18. R. P. Prickett, E. Mayer, and J. Hermel, “Waterhammer in Spacecraft Propellant
Feed Systems,” Journal of Propulsion and Power, Vol. 8, No. 3, May-June 1992.

CHAPTER 7

LIQUID PROPELLANTS

Section 6.1 gave the classification of liquid propellants. In this chapter we
discuss properties, performance, hazards, and characteristics of selected common
liquid propellants. These characteristics affect the engine and vehicle design, test
facilities, and propellant storage and handling. Today we commonly use three
liquid bipropellant combinations. Each of these propellants will be described
further in this chapter. They are: (1) the cryogenic oxygen-hydrogen propel-
lant system, used in upper stages and sometimes booster stages of space launch
vehicles; it gives the highest specific impulse for a nontoxic propellant combi-
nation, which makes it best for high vehicle velocity missions; (2) the liquid
oxygen-hydrocarbon propellant combination, used for booster stages (and a few
second stages) of space launch vehicles; its higher average density allows a more
compact booster stage, when compared to the first combination; also, historically,
it was developed before the first combination and was originally used for ballistic
missiles; (3) not a single propellant combination, but several storable propellant
combinations, are used in large rocket engines for first and second stages of
ballistic missiles and in almost all bipropellant low-thrust, auxiliary or reaction
control rocket engines (this term is defined below); they allow long-term storage
and almost instant readiness to start without the delays and precautions that come
with cryogenic propellants. There also were fuel mixtures of several storable
propellant such as alcohol and hydrazine. In Russia the nitric acid-hydrocarbon
combination was used in ballistic missiles many years ago. Today Russia and
China favor nitrogen tetroxide as the oxidizer-unsymmetrical dimethylhydrazine,
or UDMH, as the fuel for ballistic missiles and auxiliary engines. The United
States started with nitrogen tetroxide and a fuel mixture of 50% UDMH with 50%
hydrazine in the large engines for the Titan II missile. For auxiliary engines in
many satellites and upper stages the United States has used, and is still using, the

245

246 LIQUID PROPELLANTS

bipropellant of nitrogen tetroxide with monomethylhydrazine. The orhit maneu-
vering system of the Space Shuttle uses it. Alternatively many U.S. satellites
have used, and are still using monopropellant hydrazine for low-thrust auxiliary
engines.

In the last 30 years there has not been a truly new propellant used in a
rocket engine for an operational flight vehicle. Some new propellants, such as
hydroxyl ammonium nitrate, have been synthesized, manufactured, and ground
tested in thrust chambers in the last decade, but they have not found their way
into an operational rocket engine. Between 1942 and 1975 a number of other
propellants (not the ones commonly used today) were successfully flown; this
includes ammonia (X-15 Research test aircraft), ethyl alcohol (German V-2 or
U.S. Redstone), or aniline (WAC Corporal). They each had some disadvantages
and they are no longer used for operational flights today. The liquid oxidizer
nitrous oxide (N 2 O), also known as laughing gas, has found a new application
in a hybrid propulsion system (with a solid fuel) and it is discussed further in
Chapter 16 on hybrid propulsion systems. Most of the commonly used liquid
oxidizers for liquid propellant rocket engines are listed and described briefly in
this chapter.

A comparative listing of various performance quantities for a number of pro-
pellant combinations is given in Table 5-5 and in Ref. 7-1. Some important
physical properties of selected common propellants are given in Table 7-1. For
comparison water is also listed. Specific gravities and vapor pressures are shown
in Figs. 7-1 and 7-2.

7.1. PROPELLANT PROPERTIES

It is important to distinguish between the characteristics and properties of the
liquid propellants (the fuel and oxidizer liquids in their unreacted condition) and
those of the hot gas mixture, which results from the reaction in the combustion
chamber. The chemical nature of the liquid propellants and their mixture ratio
determines the properties and characteristics of both of these types. Unfortunately,
none of the practical, known propellants have all the desirable properties, and
the selection of the propellant combination is usually a compromise of various
factors, such as those listed below.

Economic Factors

Availability in quantity and a low cost are very important considerations in the
selection of a propellant. In military applications, consideration has to be given to
logistics of production, supply, storage, and other factors. The production process
should be simple, requiring only ordinary chemical equipment and available raw
materials. It is usually more expensive to use a toxic or cryogenic propellant than
a storable, nontoxic one, because it requires additional steps in the operation, more
safety provisions, additional design features, longer check-out procedures prior
to launch, and often better trained personnel.

Temperature, K

100 150 200 250 300 350

247

FIGURE 7-1. Specific gravities of several liquid propellants as a function of temperature.

248 LIQUID PROPELLANTS

TABLE 7-1. Physical Properties of Several Common Liquid Propellants

Propellant

Liquid

Fluorine

Hydrazine

Liquid

Hydrogen

Methane

Monomethyl-

hydrazine

Chemical formula

N 2 H 4

H 2

CH 4

CH 3 NHNH 2

Molecular mass

37.997

32.045

2.016

16.03

46.072

Melting or freezing

53.54

275.16

14.0

90.5

220.7

point (K)

Boiling point (K)

85.02

387.46

20.4

111.6

360.8

Heat of vaporization

172.2''

1219''

446

510''

808

(kJ/kg)

Specific heat

0.368

0.736

1.75''

0.835''

0.700

(kcal/kg-K)

(85 K)

(293 K)

(20.4 K)

(298 K)

0.357

0.758

—

0.735

Specific gravity"

(69.3 K)
1.636

(338 K)
1.0037

0.071

0.424

(393 K)
0.8702

(66 K)

(298 K)

(20.4 K)

(111.5 K)

(298 K)

1.440

0.952

0.076

—

0.857

Viscosity

(93 K)
0.305

(350 K)
0.97

(14 K)
0.024

0.12

(311 K)
0.775

(centipoise)

(77.6 K)

(298 K)

(14.3 K)

(111.6 K)

(298 K)

0.397

0.913

0.013

0.22

0.40

(70 K)

(330 K)

(20.4 K)

(90.5 K)

(344 K)

Vapor pressure

0.037

0.0019

0.2026

0.033

0.0066

(MPa)

(77 K)

(298 K)

(23 K)

(100 K)

(298 K)

0.0065

0.016

0.87

0.101

0.638

(66.5 K)

(340 K)

(30 K)

(117 K)

(428 K)

"Red fuming nitric acid (RFNA) has 5 to 20% dissolved NO 2 with an average molecular mass of
about 60, and a density and vapor pressure somewhat higher than those of pure nitric acid.

*At boiling point.

"Reference for specific gravity ratio: 10^ kg/m^ or 62.42 Ibm/ft^.

Performance of Propellants

The rocket engine performance can be compared on the basis of the specific
impulse, the exhaust velocity, the characteristic velocity, or other engine param-
eters. They have been explained in Chapters 3, 5, and 6. The specific impulse
and exhaust velocity are functions of pressure ratio, specific heat ratio, com-
bustion temperature, mixture ratio, and molecular mass. Values of performance
parameters for various propellant combinations can be calculated with a high
degree of accuracy and several are listed in Table 5-5. Very often the perfor-
mance is expressed in terms of fiight performance parameters for a given rocket

7.1. PROPELLANT PROPERTIES 249

Nitric Acid"
(99% pure)

Nitrogen

Tetroxide

Liquid

Oxygen

Rocket Fuel
RP-1, RP-2

Unsymmetrical

Dimethyl-

hydrazine

(UDMH)

Water

HNO 3

N 2 O 4

O 2

Hydrocarbon

CH 1.97

(CH3)2NNH2

H 2 O

63.016

92.016

31.988

~ 175

60.099

18.02

231.6

261.95

54.4

225

216

273.15

355.7

294.3

90.0

460-540

335.5

373.15

480

413*

213

246*

543

2253*

0.042

0.374

0.4

0.45

0.704

1.008

(311 K)
0.163
(373 K)

(290 K)
0.447
(360 K)

(65 K)

(298 K)

(298 K)
0.715
(340 K)

(273.15 K)

1.549

1.447

1.14

0.58

0.7861

1.002

(273.15 K)

(293 K)

(90.4 K)

(422 K)

(298 K)

(373.15 K)

1.476

1.38

1.23

0.807

0.784

1.00

(313.15 K)

(322 K)

(77.6 K)

(289 K)

(244 K)

(293.4 K)

1.45

0.47

0.87

0.75

0.492

0.284

(273 K)

(293 K)

(53.7 K)

(289 K)

(298 K)

(373.15 K)

0.33

0.19

0.21

0.48

1.000

(315 K)

(90.4 K)

(366 K)

(300 K)

(277 K)

0.0027

0.01014

0.0052

0.002

0.0223

0.00689

(273.15 K)

(293 K)

(88.7 K)

(344 K)

(298 K)

(312 K)

0.605

0.2013

0.023

0.1093

0.03447

(343 K)

(328 K)

(422 K)

(339 K)

(345 K)

application, as explained in Chapter 4. Here the average density, the total impulse,
and the engine mass ratio usually enter into a complex flight relation equation.

For high performance a high content of chemical energy per unit of propel-
lant mixture is desirable because it permits a high chamber temperature. A low
molecular mass of the product gases of the propellant combination is also desir-
able. It can be accomplished by using fuels rich in combined hydrogen, which is
liberated to become gas during the reaction. A low molecular mass is obtained if
a large portion of the hydrogen gas produced does not combine with oxygen. In
general, therefore, the best mixture ratio for many bipropellants is not necessar-
ily the stoichiometric one (which results in complete oxidation and yields a high
flame temperature) but usually a fuel-rich mixture containing a large portion of
low-molecular-mass reaction products, as shown in Chapter 5.

Vapor pressure, atm

Temperature, K

0 100 200 300 400 500

250

FIGURE 7-2. Vapor pressures of several liquid propellants as a function of temperature.

7.1. PROPELLANT PROPERTIES 251

If very small metallic fuel particles of beryllium or aluminum are suspended
in the liquid fuel, it is theoretically possible to increase the specific impulse
by between 9 and 18%, depending on the particular propellant combination, its
mixture ratio, and the metal powder additive. Gelled propellants with suspended
small solid particles have been ground tested successfully with storable fuels.
For gelled propellants, see Section 7.5.

The chemical propellant combination that has the highest potential specific
impulse (approximately 480 sec at 1000 psia chamber pressure and expansion
to sea-level atmosphere, and 565 sec in a vacuum with a nozzle area ratio of
50) uses a toxic liquid fluorine oxidizer with hydrogen fuel plus suspended toxic
solid particles of beryllium; as yet a practical means for storing these propellants
and a practical rocket engine have not been developed.

Common Physical Hazards

Although the several categories of hazards are described below, they do not all
apply to every propellant or every bipropellant combination. The hazards are
different for each specific propellant and must be carefully understood before
working with that propellant. The consequences of unsafe operation or unsafe
design are usually also unique to several propellants.

Corrosion. Various propellants, such as nitrogen tetroxide, nitric acid, or hydro-
gen peroxide, have to be handled in containers and pipelines of special materials.
If the propellant were permitted to become contaminated with corrosion prod-
ucts, its physical and chemical properties could change sufficiently to make it
unsuitable for rocket operation. The corrosion of the gaseous reaction products
is important in applications in which the reaction products are likely to damage
structure and parts of the vehicle or affect communities and housing near a test
facility or launch site.

Explosion Hazard. Some propellants, such as hydrogen peroxide and
nitromethane, are unstable and tend to detonate under certain conditions of impu-
rities, temperature, and shock. If liquid oxidizers (e.g., liquid oxygen) and fuels
are mixed together, they can be detonated. Unusual, rare flight vehicle launch or
transport accidents have caused such mixing to occur (see Refs. 7-2 and 7-3).

Fire Hazard. Many oxidizers will start chemical reactions with a large variety of
organic compounds. Nitric acid, nitrogen tetroxide, fluorine, or hydrogen peroxide
react spontaneously with many organic substances, resulting in a conflagration.
Most of the fuels are readily ignitable when exposed to air and heat. Also some
household dusts, certain paints, or smoke particles can be oxidized.

Accidental Spills. Unforeseen mishaps during engine operation and traffic acci-
dents on highways or railroads while transporting hazardous materials, including
propellants, have on occasion caused spills, which expose people to unexpected

252 LIQUID PROPELLANTS

fires or potential health hazards. The U.S. Department of Transportation has
rules for marking and containing hazardous materials during transport and also
guidelines for emergency action (see Ref. 7-4).

Health Hazards. Many propellants are toxic or poisonous, and special pre-
cautions have to be taken to protect personnel. Fluorine, for example, is very
poisonous. Toxic propellant chemicals or poisonous exhaust species can enter
the human body in several ways. The resulting health disorders are propellant
specific. Nitric acid can cause severe skin burns and tissue disintegration. Skin
contact with aniline or hydrazine can cause nausea and other adverse health
effects. Hydrazine, monomethylhydrazine, unsymmetrical dimethylhydrazine, or
hydrazine hydrate are known animal carcinogens (cancer-causing substances)
and suspected human carcinogens. Many propellant vapors cause eye irritation,
even in very small concentration. Inadvertent ingestion (swallowing) of many
propellants can also cause severe health degradation.

The inhalation of certain toxic exhaust gases or gaseous or vaporized propel-
lants is perhaps the most common health hazard. It can cause severe damage
if the exposure is for long duration or in concentrations that exceed established
maximum threshold values. In the United States the Occupational Safety and
Health Administration (OSHA) has established limits or thresholds on the allow-
able exposure and concentration for most propellant chemicals. Several of these
threshold limits are mentioned later in this chapter. Toxic gases in the exhaust
could include hydrofluoric acid (HF) gas; its OSHA 8-hr personnel exposure
limit is 3 ppm (volumetric parts per million) and its short-term (typically, 15
min) exposure limit is 6 ppm. A concentration of 3000 ppm or 0.3% can be fatal
within a few seconds. References 7-2 and 7-5 give more information on toxic
effects.

The corrosion, explosion, and fire hazards of many propellants put severe
limitations on the materials, the handling, and the design of rocket-propelled
vehicles and their vehicle’s engine compartments. Not only is the rocket system
itself exposed to the hazardous propellant, but adjacent personnel, structural parts,
electrical and other vehicle equipment, and test and launch facilities have to
be properly protected against the effects of possible leaks, fumes, and fires or
explosions from propellant accumulations.

Material Compatibility. Several liquid propellants have only a limited num-
ber of truly compatible materials, both metals and nonmetals, such as gaskets or
0-rings. There have been unfortunate failures (causing fires, leakage, corrosion,
or malfunctions) when an improper or incompatible material was used in the
hardware of a rocket engine. Depending on the specific component and load-
ing conditions, these structural materials have to withstand high stresses, stress
corrosion, and in some versions high temperatures or abrasion. Several specific
material limitations are mentioned in the next section. Certain materials catalyze
a self-decomposition of stored hydrogen peroxide into water and oxygen, mak-
ing long-term storage difficult and, if confined, causing its closed container to

7.1. PROPELLANT PROPERTIES 253

explode. Many structural materials, when exposed to cold, cryogenic propellants,
can become very brittle.

Desirable Physical Properties

Low Freezing Point. This permits operation of rockets in cold weather. The
addition of small amounts of freezing point depressant has been found to help
lower the freezing point of some liquid propellants which otherwise might solidify
at relatively high temperature.

High Specific Gravity. A denser propellant will provide a larger mass of pro-
pellants in a given vehicle tank volume. It permits a smaller tank volume and,
consequently, a lower structural vehicle mass and lower aerodynamic drag. Pro-
pellant specific gravity, therefore, has an important effect on the maximum flight
velocity and range of any rocket-powered vehicle or missile, particularly when
flying within the Earth’s atmosphere, as explained in Chapter 4. Specific gravities
for various propellants are plotted in Fig. 7-1. A variation of the temperature of
stored propellant will cause a change of the liquid level in the tank.

For any given mixture ratio r, the average specific gravity of a propellant
combination Sav can be determined from the specific gravities of the fuel Sf and of
the oxidizer So ■ The average specific gravity is defined as the mass of the fuel and
oxidizer, divided by the sum of their volumes. Here the mixture ratio is defined
as the oxidizer mass flow rate divided by the fuel mass flow rate. (Eq. 6-1):

^ So8f(l+r)

rSf + So

(7-1)

Values of 5av for various propellant combinations are listed in Table 5-5. The
value of 5av can be increased by adding heavy materials to the propellants, either
by solution or colloidal suspension. The identical type of equation can be written
for the average density pav in terms of the fuel density and the oxidizer density:

Pav —

PoPfjl + r)
Pfr + Po

(7-2)

In the SI system of units the specific gravity has the same numerical value
as the density expressed in units of grams per cubic centimeter or kg/liter. In
some performance comparisons the parameter density specific impulse Id is used.
It is defined as the product of the average specific gravity S and the specific
impulse 4:

Id = &..Is (7-3)

Stability. No deterioration and no decomposition with long-term (over 15 years)
storage and minimal reaction with the atmosphere have been attained with many
propellants. Good chemical stability means no decomposition of the liquid pro-
pellant during operation or storage, even at elevated temperature. A good liquid

254 LIQUID PROPELLANTS

propellant should also have no chemical deterioration when in contact with tub-
ing pipes, tank walls, valve seats, and gasket materials, even at relatively high
ambient temperatures. No appreciable absorption of moisture and no adverse
effects of small amounts of impurities are desirable properties. There should be
no chemical deterioration when liquid flows through the hot cooling jacket pas-
sages of a regeneratively cooled thrust chamber. Some hydrocarbons (e.g., some
olefins) decompose and form carbonaceous deposits on the hot inside surfaces
of the cooling passage. These deposits can become hard, reduce the heat flow,
increase the local metal temperatures, and thus can cause the metal to weaken and
fail. About 1% per year of stored 90% concentrated hydrogen peroxide decom-
poses even in clean storage tanks. Between 1 and 20% of a cryogenic propellant
(stored in a flight vehicle) evaporates every day in an insulated tank.

Heat Transfer Properties. High specific heat, high thermal conductivity, and
a high boiling or decomposition temperature are desirable for propellants that are
used for thrust chamber cooling (see Section 8.5).

Pumping Properties. A low vapor pressure permits not only easier handling
of the propellants, but also a more effective pump design in applications where
the propellant is pumped. This reduces the potential for cavitation, as explained in
Chapter 10. If the viscosity of the propellant is too high, then pumping and engine-
system calibration become difficult. Propellants with high vapor pressure, such as
liquid oxygen, liquid hydrogen, and other liquefied gases, require special design
provisions, unusual handling techniques, and special low-temperature materials.

Temperature Variation of Physical Properties. The temperature variation
of the physical properties of the liquid propellant should be small and should
be very similar for the fuel and for the oxidizer. For example, a wide tempera-
ture variation in vapor pressure and density (thermal coefficient of expansion) or
an unduly high change in viscosity with temperature makes it difficult to accu-
rately calibrate a rocket engine flow system or predict its performance over any
reasonable range of operating temperatures.

Ignition, Combustion, and Flame Properties

If the propellant combination is spontaneously ignitable, it does not require an
ignition system. This means that burning is initiated as soon as the oxidizer and
the fuel come in contact with each other. Spontaneously or self-ignitable propel-
lant combinations are often termed hypergolic propellants. Although an ignition
system is not a very objectionable feature, its elimination is usually desirable
because it simplifies the propulsion system. All rocket propellants should be read-
ily ignitable and have only a very short ignition time delay in order to reduce
the potential explosion hazard during starting. Starting and ignition problems are
discussed further in Section 8.6.

Nonspontaneously ignitable propellants have to be heated by external means
before ignition can begin. Igniters are devices that accomplish an initial slight

7.1. PROPELLANT PROPERTIES 255

pressurization of the chamber and the initial heating of the propellant mixture
to the point where steady flow combustion can be self-sustained. The amount
of energy added by the igniter to activate the propellants should be small so
that low-power ignition systems can be used. The energy required for satis-
factory ignition usually diminishes for increasing ambient temperature of the
propellant. At low ambient temperatures ignition can be very slow (0.05 to
0.02 sec).

Certain propellant combinations bum very smoothly without combustion vibra-
tion which are gas pressure oscillations. Other propellant combinations do not
demonstrate this combustion stability and, therefore, are less desirable. Combus-
tion is treated in Chapter 9.

Smoke formation is objectionable in many applications because of the smoke
deposits on the surrounding equipment and parts. Smoke and brilliantly luminous
exhaust flames are objectionable in certain military applications because they
can be easily detected. In some applications the condensed species in the exhaust
gas can cause surface contamination on spacecraft windows or optical lenses,
and the electrons in the flame can cause undesirable interference or attenuation
of communications radio signals . See Chapter 20 for information on exhaust
plumes.

Property Variations and Specifications

The propellant properties and quality must not vary from batch to batch, because
this can affect engine performance, combustion, and physical or chemical prop-
erties. The same propellant must have the same composition, properties, and
storage or rocket operating characteristics if manufactured at different times or
if made by different manufacturers. For these reasons propellants are purchased
against specifications which define ingredients, maximum allowable impurities,
packaging methods or compatible materials, allowable tolerances on physical
properties (such as density, boiling point, freezing point, viscosity, or vapor
pressure), quality control requirements, cleaning procedures for containers, doc-
umentation of inspections, laboratory analyses, or test results. A careful chemical
analysis of the composition and impurities is necessary. Reference 7-6 describes
some of these methods of analysis.

Additives

Altering and tailoring propellant properties can be achieved with additives. For
example, to make a nonhypergolic fuel become hypergolic (readily ignited), a
reactive ingredient has been added. To desensitize concentrated hydrogen perox-
ide and reduce self-decomposition, it is diluted with 3 to 15% of clean water.
To increase density or to alleviate certain combustion instabilities, a fine pow-
der of a heavy solid material, such as aluminum, can be suspended in the fuel.
The use of additives to lower the freezing point temperature has already been
mentioned.

256 LIQUID PROPELLANTS

7.2. LIQUID OXIDIZERS

The most energetic oxidizer with the highest specific impulse is liquid fluorine.
It has a high density and is extremely corrosive and toxic. It has been tested in
several complete rocket engines but has been abandoned because of its hazards.
Many different types of new storable and cryogenic liquid oxidizer propellants
have been synthesized, tested in small thrust chambers, or proposed. For high spe-
cific impulse this includes boron-oxygen-fluorine compounds, oxygen-fluorine
compounds, nitrogen-fluorine formulations, and fluorinated hydrocarbons; how-
ever, they all have some undesirable characteristics and these synthetic oxidizers
have not been proven to be practical. Oxidizer liquids that have been used in
experimental liquid rocket engines include mixtures of liquid oxygen and liq-
uid fluorine, oxygen difluoride (OF 2 ), chlorine trifluoride (CIF 3 ), or chlorine
pentafluoride (CIF 5 ). All of these are highly toxic, very corrosive, and are not
used today. Several commonly used oxidizers are listed below.

Liquid Oxygen (O 2 )

Liquid oxygen, often abbreviated as LOX, boils at 90 K at atmospheric pressure;
at these conditions it has a specific gravity of 1.14 and a heat of vaporization
of 213 kJ/kg. It is widely used as an oxidizer and burns with a bright white-
yellow flame with most hydrocarbon fuels. It has been used in combination
with alcohols, jet fuels (kerosene-type), gasoline, and hydrogen. As shown in
Table 5-5, the attainable performance is relatively high, and liquid oxygen is
therefore a desirable and commonly used propellant in large rocket engines. The
following missiles and space launch vehicles use oxygen: (1) with jet fuel or RP-
1 (kerosene) — Atlas V and Soyuz (Russia); (2) with hydrogen — Space Shuttle,
Ariane-V (France), Delta IV, and Centaur upper stage; (3) with alcohol — some
amateur rocket engines. Figs. 1-4 and 6-1 show units that use oxygen. Figures
5-1 to 5-6 give theoretical performance data for liquid oxygen with a kerosene-
type fuel.

Although it usually does not burn spontaneously with organic materials at
ambient pressures, combustion or explosions can occur when a confined mix-
ture of oxygen and organic matter is suddenly pressurized. Impact tests show
that mixtures of liquid oxygen with many commercial oils or organic materials
will detonate. Liquid oxygen supports and accelerates the combustion of other
materials. Handling and storage are safe when contact materials are clean. Liquid
oxygen is a noncorrosive and nontoxic liquid and will not cause the deteriora-
tion of clean container walls. When in prolonged contact with human skin, the
cryogenic propellant causes severe frostbite. Because liquid oxygen evaporates
rapidly, it cannot be stored readily for any great length of time. If liquid oxygen
is used in large quantities, it is often produced very close to its geographical
point of application. Liquid oxygen can be obtained in several ways, such as by
fractionated distillation of liquid nitrogen out of liquid air.

It is necessary to insulate all lines, tanks, valves, and so on that contain liquid
oxygen in order to reduce the evaporation loss. Rocket propulsion systems which

7.2. LIQUID OXIDIZERS 257

remain filled with liquid oxygen for several hours and liquid oxygen storage
systems have to be well insulated against absorbing heat from the surroundings.
External drainage provisions have to be made on all liquid oxygen tanks and
lines to eliminate the water that condenses on the walls.

Example 7-1. Estimate the approximate temperature and volume change of liquid oxy-
gen if an oxygen tank is pressurized to 8.0 atm for a long time before engine start. Assume
the tank is 60% full and the evaporated oxygen is caught, refrigerated, and recondensed
(constant mass).

SOLUTION. Using Table 7-1 and Figs. 7-1 and 7-2, the vapor pressure goes from 1.0
atm (0.1 MPa) to 8 atm (about 0.8 MPa) and the equilibrium temperature goes from the
boiling point of 90 K at 1.0 atm to about 133 K at 8 atm. The corresponding specific
gravities are 1.14 and 0.88, respectively. This is an increase of 1.14/0.88 = 1.29. (29%
more volume). The tank would be 77% full.

In tanks with turbopump feed systems the actual tank pressures are lower (typically 2 to
4 atm) and the evaporated oxygen is vented, causing a cooling effect on the liquid surface.
So the numbers calculated above are too large (8 atm was selected to clearly show the
effect). The warming occurs when there is a long hold period of a pressurized cryogenic
propellant tank and is most pronounced when the final portion of the propellant is being
emptied. Nevertheless the higher temperature, higher vapor pressure, and lower density
can cause changes in mixture ratio, required tank volume, and pump suction condition
(see Section 10.5). Therefore tanks with cryogenic propellant are insulated (to minimize
heat transfer and density changes) and are pressurized only shortly before engine start, so
as to keep the propellant at its lowest possible temperature and highest density.

Hydrogen Peroxide (H2O2)

This propellant is not only a powerful liquid oxidizer but also a clean-burning
monopropellant. In rocket applications, hydrogen peroxide has been used in a
highly concentrated form of 70 to 99%; the remainder is mostly water. Com-
mercial peroxide is approximately 30% concentrated. For many paper and pulp
mills it is 70% concentrated hydrogen peroxide. It was used in gas generators
and rocket applications between 1938 and 1965 (X-1 and X-15 research aircraft).
Today 90% hydrogen peroxide is the most common.

As a monopropellant it decomposes according to the following chemical reac-
tion, forming superheated steam and gaseous oxygen:

H 2 O 2 ^ H 2 O + j 02 + heat

This decomposition is brought about by the action of catalysts such as silver
screens, various liquid permanganates, solid manganese dioxide, platinum, and
iron oxide. In fact, most impurities act as a catalyst. H 2 O 2 is hypergolic with
hydrazine and will burn well with kerosene. The theoretical specific impulse of
90% hydrogen peroxide is 154 sec, when used as a monopropellant with a solid
catalyst bed.

258 LIQUID PROPELLANTS

Even under favorable conditions H2O2 will often decompose at a slow rate
during storage, about 1% per year for 90 to 95%, and gas will bubble out of the
liquid. Contaminated liquid peroxide must be disposed of before it reaches about
448 K, when an explosion may occur. Concentrated peroxide causes severe bums
when in contact with human skin and may ignite and cause fires when in contact
with wood, oils, and many other organic materials. In the past rocket engines
with hydrogen peroxide as the oxidizer have been used for aircraft boost (German
Me 163, and U.S. F 104) and a satellite launcher (Britain: Black Arrow). As a
monopropellant it was used for attitude control thmsters and for gas generators.
It has not been used for a long time, partly because of its long-term storage
stability. However, there has been some improvement and some renewed interest
in this dense oxidizer, which produces a nontoxic exhaust.

Nitric Acid (HNO3)

There are several types of nitric acid mixtures that have been used as oxidiz-
ers between 1940 and 1965; they are not used extensively today in the United
States. The most common type, red fuming nitric acid (RENA), consists of con-
centrated nitric acid (HNO3) that contains between 5 and 27% dissolved nitrogen
dioxide (NO2). The evaporating red-brown fumes are exceedingly annoying and
poisonous. Compared to concentrated nitric acid (also called white fuming nitric
acid), RENA is more energetic, more stable in storage, and slightly less corrosive
to many tank materials.

Nitric acid is highly corrosive. Only certain types of stainless steel, gold, and
a few other materials are satisfactory as storage containers or tubing materi-
als. A small addition of fluoride ion (less than 1% of hydrofluoric acid, or HE)
inhibits the nitric acid, causes a protective fluoride layer to form on the wall, and
reduces the corrosion with many metals. It is called inhibited red fuming nitric
acid (IRENA). Even with an inhibitor, nitric acid reacts with the wall materials
and forms dissolved nitrates and sometime insoluble nitrates. This changes the
properties of the acid and may cause blocking of valve and injector orifices. In
case of accidental spilling, the acid should be diluted with water or chemically
deactivated. Lime and alkali metal hydroxides and carbonates are common neu-
tralizing agents. However, nitrates formed by the neutralization are also oxidizing
agents and must be handled accordingly.

Nitric acid has been used with gasoline, various amines, hydrazine, dimethyl-
hydrazine, and alcohols. It ignites spontaneously with hydrazine, furfuryl alcohol,
aniline, and other amines. The specific gravity of nitric acid varies from 1.5 to
1.6, depending on the percentages of nitrogen dioxide, water, and impurities.
This high density permits compact vehicle construction.

Vapors from nitric acid or red fuming nitric acid have an OSHA 8-hr personnel
exposure limit or a threshold work allowance of 2 ppm (parts per million or about
5 mg/m^) and a short-term exposure limit of 4 ppm. Droplets on the skin cause
burns and sores which do not heal readily.

7.3. LIQUID FUELS 259

Nitrogen Tetroxide (N2O4), Often Abbreviated as NTO

This is a high-density yellow-brown liquid (specific gravity of 1.44). It is com-
monly used with the hypergolic fuels hydrazine, monomethylhydrazine (MMH),
and unsymmetrical dimethylhydrazine (UDMH). Although it is the most com-
mon storable oxidizer today, its liquid temperature range is narrow and it is easily
frozen or vaporized. It is only mildly corrosive when pure, but forms strong acids
when moist or allowed to mix with water. It readily absorbs moisture from the
air. It can be stored indefinitely in sealed containers made of compatible material.
It is hypergolic with many fuels and can cause spontaneous ignition with many
common materials, such as paper, grease, leather, or wood. The NO2 fumes are
reddish brown and are extremely toxic. Because of its high vapor pressure it must
be kept in relatively heavy tanks. The freezing point of N2O4 can be lowered (for
example, by adding a small amount of nitric oxide or NO) but at the penalty of
a higher vapor pressure and slightly reduced performance. This mixture of NO
and N2O4 is called mixed oxides of nitrogen (MON) and different grades have
between 2 and 30% by weight NO content.

Nitrogen tetroxide oxidizer is used with UDMH in many Russian engines and
almost all their small thrusters. It is also used with monomethylhydrazine fuel in
the Space Shuttle orbital maneuver system, multithruster reaction control system,
and in many U.S. spacecraft propulsion systems. In many of these applications
care must be taken to avoid freezing this nitrogen tetroxide. The OSHA 8-hr
personnel exposure limit is 5 ppm or 9 mg/vp? .

7.3. LIQUID FUELS

Again, many different chemicals have been proposed, investigated, and tested
as fuels. Only a few have been used in production rocket engines. Liquid fuels
other than those listed in this chapter have been used in experimental rocket
engines, in older experimental designs, and in some older production engines.
These include aniline, furfuryl alcohol, xylidine, gasoline, hydrazine hydrate,
borohydrides, -methyl and/or ethyl alcohol, ammonia, and mixtures of some of
these.

Hydrocarbon Fuels

Petroleum derivatives encompass a large variety of different hydrocarbon chem-
icals, most of which can be used as a rocket fuel. Most common are those types
that are in use with other applications and engines, such as gasoline, kerosene,
diesel oil, and turbojet fuel. Their physical properties and chemical composi-
tion vary widely with the type of crude oil from which they were refined, with
the chemical process used in their production, and with the accuracy of control
exercised in their manufacture. Typical values are listed in Table 7-2.

In general, these petroleum fuels form yellow-white, brilliantly radiating
flames and give good performance. They are relatively easy to handle, and there

260 LIQUID PROPELLANTS

TABLE 7-2. Properties of Some Typical Hydrocarbon Fuels Made from Petroleum

Jet

Fuel

Kerosene

Aviation

Gasoline

100/130

RP-1

RP-2

Specific gravity at

0.78

0.81

0.73

0.80-0.815

289 K

Freezing point (K)

213 (max.)

230

213

213 (max.)

213 (max)

Viscosity at 289 K

1.4

1.6

0.5

16.5 (at 239 K)

(cP)

Flash point" (K)

269

331

244

333

ASTM distillation
(K)

10% evaporated

347

—

337

458-483

50% evaporated

444

—

363

—

90% evaporated

511

—

391

—

Reid vapor

2 to 3

Below 1

—

pressure (psia)

Specific heat

0.50

0.49

0.53

0.50

(cal/kg-K)

Average molecular

130

175

—

mass (g/mol)

Sulfur, total mg/kg

—

30 (max)

1 (max)

"Tag closed cup method.

is an ample supply of these fuels available at relatively low cost. A specifically
refined petroleum product particularly suitable as a rocket propellant has been
designated RP-1 (Rocket Propellant number 1). It is basically a kerosene-type
fuel mixture of hydrocarbons with a somewhat narrow range of densities and
vapor pressure. Several hydrocarbon fuels can form carbon deposits on the inside
of cooling passages, impeding the heat transfer and raising wall temperatures.
Ref. 7-7 indicates that this carbon formation depends on fuel temperature in
the cooling jacket, the particular fuel, the heat transfer, and the chamber wall
material. RP-1 is low in olefins and aromatics, which can cause carbonaceous
deposits inside fuel cooling passages. RP-1 has been used with liquid oxygen
in many early rocket engines (see Figs. 5-1 to 5-6 and Ref. 7-8). A very
similar kerosene-type fuel is used in Russia today. A few years ago RP-2 was
substituted for RP-1 in U.S. rocket applications. The principal difference is a
reduced sulphur content (see Table 7-2), because this impurity was believed to
have caused corrosion in cooling jackets of certain thrust chambers. See Ref.
7-8.

Methane (CH 4 ) is a cryogenic hydrocarbon fuel. It is denser than liquid hydro-
gen and relatively low in cost. Compared to petroleum-refined hydrocarbons, it
has highly reproducible properties. With liquid oxygen it is a candidate pro-
pellant combination for launch vehicle booster rocket engines and also reaction

7.3. LIQUID FUELS 261

control engines when oxygen is available from the main engines. Experimental
oxygen -methane engines have been tested, but they have not yet flown.

Liquid Hydrogen (H2)

Liquid hydrogen, when burned with liquid fluorine or liquid oxygen, gives a high
performance, as shown in Table 5-5. It also is an excellent regenerative coolant.
With oxygen it burns with a colorless flame; however, the shock waves in the
plume may be visible. Of all known fuels, liquid hydrogen is the lightest and
the coldest, having a specific gravity of 0.07 and a boiling point of about 20 K.
The very low fuel density requires bulky fuel tanks, which necessitate very large
vehicle volumes. The extremely low temperature limits the selection of materials
for pumps, cooling jackets, tanks, and piping because many metals become brittle
at low temperatures.

Because of its low temperature, liquid hydrogen tanks and lines have to be
well insulated to minimize the evaporation of hydrogen or the condensation of
moisture or air on the outside with the subsequent formation of liquid or solid air
or ice. A vacuum jacket often has been used in addition to insulating materials.
All common liquids and gases solidify in liquid hydrogen. These solid particles in
turn plug orifices and valves. Therefore, care must be taken to purge all lines and
tanks of air and moisture (flush with helium or pull vacuum) before introducing
the propellant. Mixtures of liquid hydrogen and solid oxygen or solid air can be
explosive.

Liquid hydrogen has two isomers, namely, orthohydrogen and parahydrogen,
which differ in the orientation of their nuclear spin state. As hydrogen is liquefied
and cooled to lower and lower temperatures, the relative equilibrium composition
of ortho- and parahydrogen changes. The transformation from orthohydrogen to
parahydrogen is exothermic and results in excessive boiloff, unless complete
conversion to parahydrogen is achieved during liquefaction. Liquid hydrogen is
manufactured from gaseous hydrogen by successive compression, cooling, and
expansion processes.

Hydrogen gas, when mixed with air, is highly flammable and explosive over
a wide range of mixture ratios. To avoid this danger, escaping excess hydrogen
gas (at tank vent lines) is often intentionally ignited and burned or “flared” in
air. Liquid hydrogen is used with liquid oxygen in the Delta IV launcher. Cen-
taur upper stage, the Space Shuttle main engine, and upper stage space engines
developed in Japan, Russia, Europe, India, and China.

Hydrogen burning with oxygen forms a nontoxic almost invisible exhaust
gas. This propellant combination has been applied successfully to space launch
vehicles because of its high specific impulse. Here the payload capability usually
increases greatly for relatively small increases in specific impulse. However, the
low density of hydrogen makes for a very large fuel tank, a large vehicle, and a
relatively high drag.

Some studies have shown that, when burned with liquid oxygen, a hydrocarbon
(such as methane or RP-1) can give a small advantage in space launch vehicle first

262 LIQUID PROPELLANTS

stages. Here the higher average propellant density allows a smaller vehicle with
lower drag, which compensates for the lower specific impulse of the hydrocarbon
when compared to a hydrogen fuel. Also, there are some concepts for operating
the booster-stage rocket engine initially with hydrocarbon fuel and then switching
during flight to hydrogen fuel. Engines using LOX and these two fuels, namely
hydrocarbon and hydrogen, have been called tripropellant engines. They have
not yet been fully developed or flown. Some work on experimental engines was
done in Russia, but there is no known current effort.

Hydrazine (N2H4)

Reference 7-9 gives a good discussion of this propellant, which is used as a
bipropellant fuel as well as a monopropellant. Hydrazine, monomethylhydrazine
(MMH), and unsymmetrical dimethylhydrazine (UDMH) have similar physical
and thermochemical properties. Hydrazine is a toxic, colorless liquid with a high
freezing point (274.3 K or 34°F). Hydrazine tanks, pipes, injectors, catalysts, and
valves are usually electrically heated to prevent freezing in cool weather or in
outer space. Hydrazine has a short ignition delay and is spontaneously ignitable
with nitric acid, nitrogen tetroxide, and concentrated hydrogen peroxide.

Its vapors may form explosive mixtures with air. If hydrazine is spilled on a
porous surface or a cloth, a spontaneous ignition with air can occur.

Pure anhydrous hydrazine is a stable liquid; it has been safely heated above 480
K. It has been stored in sealed tanks for over 15 years. With impurities or at higher
temperatures it decomposes and releases energy. Under pressure shock (blast
wave or adiabatic compression) hydrazine vapor or hydrazine mist can decom-
pose at temperatures as low as 367 K. Under some conditions this decomposition
can be a violent detonation, and this has caused problems in cooling passages
of experimental injectors and thrust chambers. Harmful effects to personnel may
result from ingestion, inhalation of vapors, or prolonged contact with skin. The
American Conference of Government Industrial Hygienists (ACGIH) recom-
mended 8-hr personnel exposure limit is 0.01 ppm or 0.013 mg/m^. Hydrazine
is a known animal carcinogen and a suspected carcinogen for people.

Hydrazine reacts with many materials, and care must be exercised to avoid
storage contact with materials that cause a decomposition (see Refs. 7-9 and
7-10). Tanks, pipes, injectors, catalysts, or valves must be cleaned and free of
impurities. Compatible materials include certain stainless steels (303, 304, 321,
or 347), nickel, and 1100 and 3003 series of aluminum. Iron, copper, and its
alloys (such as brass or bronze), monel, magnesium, zinc, and some types of
aluminum alloy must be avoided.

Unsymmetrical Dimethylhydrazine [(CHs)2NNH2]

A derivative of hydrazine, namely UDMH, is often used instead of or in mix-
tures with hydrazine because it forms a more stable liquid, particularly at higher
temperatures. Furthermore, it has a lower freezing point (215.9 K) and a lower

7.4. LIQUID MONOPROPELLANTS 263

boiling point (336.5 K) than hydrazine. When UDMH is burned with an oxidizer
it gives only slightly lower values of 4 than pure hydrazine. UDMH is often used
when mixed with 30 to 50% hydrazine or with 25% hydrazine hydrate. UDMH
is used in Russian and Chinese rocket engines and small thrusters. The ACGIH
recommended 8 -hr personnel exposure limit for vapor is 0.01 ppm, and UDMH
is a suspected carcinogen.

Monomethylhydrazine (CH 3 NHNH 2 )

This fuel abbreviated as MMH has been used extensively in U.S. spacecraft
rocket engines, particularly in small attitude control engines, usually with N 2 O 4
as the oxidizer. It has a better shock resistance to blast waves and a better
liquid temperature range than pure hydrazine. Like hydrazine, its vapors are eas-
ily ignited in air; the flammability limits are from 2.5 to 98% by volume at
atmospheric sea-level pressure and ambient temperature. The materials compat-
ible with hydrazine are also compatible with MMH. The specific impulse with
storable oxidizers usually is 1 or 2% lower with MMH than with N 2 H 4 .

Both MMH and UDMH are soluble in many hydrocarbons; hydrazine is not.
All hydrazines are toxic materials, but MMH is the most toxic when inhaled.
Atmospheric concentrations of all hydrazines should be kept below 0.1 ppm for
long periods of personnel exposure.

Monomethylhydrazine, when added in relatively small quantities of 3 to 15%
to hydrazine, has a substantial quenching effect on the explosive decomposition of
hydrazine. Monomethylhydrazine decomposes at 491 K, whereas hydrazine can
explode at 369 K when subjected to certain pressure shocks. MMH is a known
animal carcinogen and the ACGIH recommended personnel 8 -hour exposure limit
is 0.01 ppm.

7.4. LIQUID MONOPROPELLANTS

The propellant-feed and control-system simplicity associated with a monopropel-
lant makes this type of propellant attractive for certain applications. Hydrazine
is being used extensively as a monopropellant in small attitude and trajectory
control rockets for the control of satellites and other spacecraft and also as a
hot gas generator. (It is discussed in the preceding section.) Other monopro-
pellants (ethylene oxide or nitromethane) were tried experimentally, but are no
longer used today. Concentrated hydrogen peroxide was used for small thrusters
between 1945 and 1965 and for monopropellant gas generation in the United
States, Russia, United Kingdom or Britain, and Germany.

Ignition of monopropellants can be achieved thermally (electrical or flame
heat) or by a catalytic material. A monopropellant must be chemically and ther-
mally stable to ensure good liquid storage properties, and yet it must be easily
decomposed and reactive to quickly give good combustion.

264 LIQUID PROPELLANTS

Hydrazine as a Monopropellant

Hydrazine is not only an excellent storable fuel but also an excellent monopro-
pellant when decomposed by a suitable solid catalyst; this catalyst often needs
to be preheated for fast startup or for extending the useful catalyst life. Iridium
on porous alumina base is an effective catalyst at room temperature. At ele-
vated temperature (about 450 K) many materials decompose hydrazine, including
iron, nickel, and cobalt. See Ref. 7-9. Different catalysts and different catalyst
configurations make the decomposition reaction go to different products, result-
ing in gases varying in composition or temperature. As a monopropellant, it
is used in gas generators or in space engine attitude control rockets. A typical

80
60
40
20

0 20 40 60 80 100

Ammonia decomposition, %

FIGURE 7-3. Operating parameters for decomposed hydrazine at the exit of a catalytic
reactor as a function of the ammonia dissociation fraction. Adapted with permission from
Ref. 7-9.

7.5. GELLED PROPELLANTS 265

hydrazine monopropellant thrust chamber, its injection patterns, and its decom-
position reaction are described in Fig. 8-14. Typical operating parameters are
shown in Fig. 7-3.

The catalytic decomposition of hydrazine can be described ideally as a two-
step process; this ignores other steps and intermediate products. First, hydrazine
(N2FI4) decomposes into gaseous ammonia (NFI3) and nitrogen (N2); this reac-
tion is highly exothermic, that is, it releases heat. Second, the ammonia decom-
poses further into nitrogen and hydrogen gases, but this reaction is endothermic
and absorbs heat. These simplified reactions can be written as

3N2H4 ^ 4(1 - x)NH 3 -h (1 -h 2 x)N 2 -h 6.rH2 (7-4)

Flere x is the degree of ammonia dissociation; it is a function of the catalyst type,
size, and geometry, the chamber pressure, and the dwell time within the catalyst
bed. Figure 7-3 shows several ideal rocket engine parameters for hydrazine
monopropellant as a function of x, the fraction of ammonia that is decomposed.
The values are for an ideal thruster at 1000 psia chamber pressure with a nozzle
area ratio of 50 expanding at high altitude. The best specific impulse is attained
when little ammonia is allowed to dissociate.

7.5. GELLED PROPELLANTS

Gelled propellants have additives that make them thixotropic materials. They
have the consistency of thick paint or jelly when at rest, but they liquefy and
flow through pipes, valves, pumps, or injectors when an adequate shear stress is
applied. They offer these advantages: (1) Small aluminum fuel particles can be
suspended in the fuels. Inert solid particles can be suspended in oxidizer liquids.
This increases propellant density, density impulse, and thus reduces the size of
tanks and vehicles. (2) There is no plugging of injector orifices or valve passages
and good flow control has been demonstrated. (3) Most gelled fuel propellants are
essentially nonflammable and will not usually sustain an open fire. (4) There is
reduced susceptibility of leakage or spill, reduced sloshing of liquids in the tanks,
and the boil-off rate is reduced. (5) Long-term storage of 10 years or more without
settling or separation of the solid particles has been demonstrated. (6) Explosions
or detonations, which happen when a vehicle accident causes liquid propellants to
become inadvertently mixed, are much less likely with gelled propellants, which
are difficult to mix. (7) Short-duration pulsing has been tested. (8) Most storable
oxidizers, a few cryogenic propellants, and most liquid storable fuels can be
gelled. (9) Explosions are much less likely when a propellant tank is penetrated
by a bullet or when a missile is exposed to an external fire or a nearby detonation.

These are some of the disadvantages: (1) There is a small decrease in specific
impulse due to dilution with a gelling agent, and less efficient atomization into
small droplets (due to high viscosity) or less efficient combustion. Eor example,
the characteristic velocity c* of oxygen-kerosene propellant is decreased by

266 LIQUID PROPELLANTS

4 to 6% when the kerosene is gelled and aluminum is suspended in the fuel.

(2) Bubhle-free loading or unloading of propellants is somewhat more complex
and may take longer. Residual propellant quantity will be slightly higher because
the thixotropic fluid layer on the walls of the tanks and pipes will be thicker.

(3) Changes in ambient temperature will cause slight changes in propellant density
and viscosity and therefore also in mixture ratio; this can result in more leftover
or residual propellant and thus in a slight reduction of available total impulse.
Suspended metals can make the exhaust plume smoky and visible.

Experimental rocket engines have shown these gelled propellants to be gen-
erally safer than ordinary liquid propellants and to have adequate performance
and operational characteristics (see Refs. 7-11 and 7-12). This makes them less
susceptible to field accidents. A variety of different organic and inorganic gelling
agents have been explored with a number of different liquid propellants.

Experimental thrust chambers and rocket engine systems have been satisfacto-
rily demonstrated with several gelled propellant combinations. One experimental
engine has flown. As far as is known, no such rocket engine has yet been put
into production or flight operation.

7.6. GASEOUS PROPELLANTS

Cold gas propellants have been used successfully for reaction control systems
(RCS) for more than 60 years. The words “cold gas” (stored at ambient temper-
ature) is distinguished from a “warm gas,” which has been heated. The engine
system is simple, consisting of one or more high-pressure gas tanks, multiple
simple nozzles (often aluminum or plastic), an electrical control valve with each
nozzle, a pressure regulator, and provisions for filling and venting the gas. The
tank size will be smaller if the tank pressures are high. Pressures are typically
between 300 and 1000 MPa (about 300 to 10,000 psi), and these gas tanks have
thick wells and are heavy.

Typical cold gas propellants and some of their properties and characteristics
are listed in Table 7-3. Nitrogen, argon, dry air, krypton and Freon 14 have
been employed in spacecraft RCS. With high-pressure hydrogen or helium as
cold gas, the specific impulse is much higher, but the densities of these gases are
much lower. This requires a much larger gas storage volume and heavier high-
pressure tanks. In most applications the extra inert mass outweighs the advantage
of better performance. In a few applications the gas (and its storage tank) are
heated electrically or chemically. This improves the specific impulse and allows
a smaller tank, but it also introduces complexity.

The selection of the gas propellant, the storage tanks, and RCS design depend
on many factors, such as volume and mass of the storage tanks, the maximum
thrust and total impulse, the gas density, required maneuvers, duty cycle, and
flight duration. Cold gas systems have been used for total impulses of perhaps
22,200 N-sec or 5000 Ibf-sec. Higher values usually employ liquid propellants.

If the operation is short (most of the gas consumed in a few minutes, while
the main engine is running), the gas expansion will be close to adiabatic (no heat

7.7. SAFETY AND ENVIRONMENTAL CONCERNS 267

TABLE 7-3. Properties of Gaseous Propellants Used for Auxiliary Propulsion

Propellant

Molecular

Mass

Density"

(lb/ft3)

Specific heat
ratio k

Theoretical Specific
Impulse* (sec)

Hydrogen

2.0

1.77

1.40

284

Helium

4.0

3.54

1.67

179

Methane

16.0

14.1

1.30

114

Nitrogen

28.0

24.7

1.40

Air

28.9

25.5

1.40

Argon

39.9

35.3

1.67

Krypton

83.8

74.1

1.63

“At 5000 psia and 20“C.

*In vacuum with nozzle area ratio of 50:1 and initial temperature of 20°C.

absorption by gas) and often is analyzed as isentropic. The temperature of the
gas, the pressure, and specific impulse will drop as the gas is consumed. For long
intermittent operations (months or years in space) the heat from the spacecraft
is transfered to the gas and the tank temperature stays essentially constant; the
expansion will be nearly isothermal.

The advantages and disadvantages of cold gas thrusters and systems are
described in Section 8.3 in the discussion of low thrust.

7.7. SAFETY AND ENVIRONMENTAL CONCERNS

To minimize the hazards and potential damage inherent in reactive propellant
materials, it is necessary to be very conscientious about the likely risks and
hazards (see Refs. 7-4, 7-13, and 7-14). This concerns toxicity, explosiveness,
fire or spill danger, and others mentioned in Section 7.1. Before an operator,
assembler, maintenance mechanic, supervisor, or engineer is allowed to transfer
or use a particular propellant, he or she should receive safety training in the
particular propellant, its characteristics, its safe handling or transfer, potential
damage to equipment or the environment, and the countermeasures for limiting
the consequences in case of an accident. They must also understand the potential
hazards to the health of personnel, first aid, remedies in case of contact exposure
of the skin, ingestion, or inhaling, and the use of safety equipment. Examples of
safety equipment are protective clothing, face shields, detectors for toxic vapors,
remote controls, warning signals, or emergency water deluge. The personnel
working with or being close to highly toxic materials usually have to undergo
periodic health monitoring. Also rocket engines need to be designed for safety
to minimize the occurrence of a leak, an accidental spill, an unexpected fire, or
other potentially unsafe conditions. Most organizations have one or more safety
specialists who review the safety of the test plans, manufacturing operations,
designs, procedures, or safety equipment. With the proper training, equipment,
precautions, and design safety features, all propellants can be handled safely.

268 LIQUID PROPELLANTS

If a safety violation occurs or if an operation, design, procedure, or practice
is found to be (or appears to be) unsafe, then a thorough investigation of the
particular item or issue should be undertaken, the cause of the lack of safety
should be investigated and identified, and an appropriate remedial action should
be selected and initiated as soon as possible.

The discharge of toxic exhaust gases to the environment and their dispersion
by the wind can cause exposure of operating personnel as well as the people in
nearby areas, and can cause damage to plants and animals. This is discussed in
Section 20.2. The dumping or spilling of toxic liquids can contaminate subter-
ranean aquifers and surface waters, and their vapors can pollute the air. Today
the type and amount of gaseous and liquid discharges are regulated and moni-
tored by government authorities. These discharge quantities must be controlled
or penalties will be assessed against violators. Obtaining a permit to discharge
can be a lengthy and involved procedure.

PROBLEMS

1 . Plot the variation of the density specific impulse (product of average specific gravity
and specific impufse) with mixture ratio and expfain the meaning of the curve. Use the
theoretical shifting specific impulse values of Fig. 5-1 and the specific gravities from
Fig. 7-1 or Table 7-1 for the liquid oxygen-RP-1 propellant combination.

Answers: Check point at r = 2.0; = 290; Id = 303; i5av = 1-01.

2 . Prepare a table comparing the relative merits of liquid oxygen and nitrogen tetroxide
as rocket engine oxidizers.

3 . Derive Eq. 7-1 for the average specific gravity.

4 . A rocket engine uses liquid oxygen and RP-1 as propellants at a design mass mixture
ratio of 2.40. The pumps used in the feed system are basically constant-volume flow
devices. The RP-1 hydrocarbon fuel has a nominal temperature of 298 K and it can
vary at about ±25°C. The liquid oxygen is nominally at its boiling point (90 K),
but, after the tank is pressurized, this temperature can increase by up to 30 K during
long time storage. What are the extreme mixture ratios under unfavorable temperature
conditions? If this engine has a nominal mass flow rate of 100 kg/sec and a duration
of 100 sec, what is the maximum residual propellant mass when the other propellant
is fully consumed? Use the curve slopes of Fig. 7-1 to estimate changes in density.
Assume that the specific impulse is constant for the relatively small changes in mixture
ratio, that small vapor pressure changes have no influence on the pump flow, that there
is no evaporation of the oxygen in the tank, and that the engine has no automatic
control for mixture ratio. Assume zero residual propellant.

5 . The vehicle stage propelled by the rocket engine in Problem 4 has a design mass ratio
nif/nio of 0.50 (see Eq. 4-6). For the specific impulse please use a value half way
between the shifting and the frozen equilibrium curves of Fig. 5-1. How much will
the worst combined changes in propellant temperatures affect the mass ratio and the
ideal gravity-free vacuum velocity?

REFERENCES 269

6. (a) What should he the approximate percent ullage volume for nitrogen tetroxide tank

when the vehicle is exposed to ambient temperatures between about 50°F and
about 150°F?

(b) What is maximum tank pressure at 150°F.

Answers: (a) 15 to 17%; the variation is due to the nonuniform temperature distribution
in the tank; (b) 6 to 7 atm; (c) vapor pressure, nitrogen monoxide content in the
oxidizer, chemical reactions with wall materials, or impurities that result in largely
insoluble gas products.

7. An insulated, long vertical, vented liquid oxygen tank has been sitting on the sea-level
launch stand for a period of time. Assume no heat loss to the tank wells. The surface
of the liquid is at atmospheric pressure and is 10.2 m above the closed outlet at the
bottom of the tank. If there is no circulation, what will be the temperature, pressure,
and density of the oxygen at the tank outlet?

SYMBOLS

III density specific impulse, sec

7j specific impulse, sec

k ratio of specific heats

r mixture ratio (mass flow rate of oxidizer to mass flow rate of fuel)

Greek Letters

^av

Pav ■: Pf ■: Po

average specific gravity of mixture
specific gravity of fuel
specific gravity of oxidizer

densities average, fuel, and oxidizer, kg/m^ (Ibm/ft^)

REFERENCES

7-1. S. F. Samer, Propellant Chemistry, Reinhold Publishing, New York, 1966.

7-2. Chemical Rocket Propellant Hazards, Vol. 1, General Safety Engineering Design
Criteria, Chemical Propulsion Information Agency (CPI A) Publication 194, Octo-
ber 1971.

7-3. L. C. Sutherland, “Scaling Law for Estimating Liquid Propellant Explosive Yields,”
Journal of Spacecraft and Rockets, March-April 1978, pp. 124-125.

7-4. NIOSH Pocket Guide to Chemical Hazards, DHHS (NIOSH) Publication No. 2005-
149, Department of Health and Human Services, Washington DC, September 2005.

7-5. 1990-1991 Threshold Limit Values for Chemical Substances and Physical Agents
and Biological Exposure Indices, American Conference of Government Industrial
Hygienists, Cincinnati, OH, 1990 (revised periodically).

270

LIQUID PROPELLANTS

7-6. H. E. Malone, The Analysis of Rocket Propellants , Academic Press, New York,
1976.

7-7. K. Liang, B. Yang, and Z. Zhang, “Investigation of Heat Transfer and Coking
Characteristics of Hydrocarbon Fuels,” Journal of Propulsion and Power, Vol. 14,
No. 5, September-October 1998.

7-8. Detail Specification — Propellant Rocket Grade Kerosene, Department of Defense,
Washington DC, M1L-DTL-25576D, 20 May 2005.

7-9. E. W. Schmidt, Hydrazine and Its Derivatives, Preparation, Properties, Applica-
tions, 2nd ed., John Wiley & Sons, New York, 2001.

7-10. O. M. Morgan and D. S. Meinhardt, “Monopropellant Selection Criteria —
Hydrazine and other Options,” AIAA Technical Paper 99-2595, June 1999.

7-11. K. F. Hodge, T. A. Crofoot, and S. Nelson, “Gelled Technical Propellants for
Tactical Missile Application,” AIAA Technical Paper 99-2976, June 1999.

7-12. B. D. Allen, “History, Development and Testing of Thixotropic Gels for Advanced
Systems,” AIAA Propulsion Conference, July 1985.

7-13. Hazards of Chemical Rockets and Propellants , Vol. I, Safety, Health and the
Environment , and Vol. Ill, Liquid Propellants , Chemical Propulsion Information
Analysis Center, CPIA Publication 394, 1984.

7-14. Emergency Response Guidebook, National Institute for Occupational Safety and
Health (NIOSH), Department of Health and Human Services, Washington, DC,
rev. ed., 2004.

CHAPTER 8

THRUST CHAMBERS

This chapter describes thrust chambers, their components, cooling, ignition, and
heat transfer. The thrust chamber is the key subassembly of a rocket engine.
Here the liquid propellants are metered, injected, atomized, vaporized, mixed,
and burned to form hot reaction gas products, which in turn are accelerated and
ejected at supersonic velocity (see Refs. 6-1, 6-2, and 8-1). A rocket thrust
chamber assembly (Figs. 8-1 and 8-2) has an injector, a combustion chamber,
a supersonic nozzle, and mounting provisions. All have to withstand the extreme
heat of combustion and the various forces, including the transmission of the
thrust force to the vehicle. There also is an ignition system if nonspontaneously
ignitable propellants are used. Some thrust chamber assemblies also have inte-
grally mounted propellant valves and sometimes a thrust vector control device,
as described in Chapter 18. Table 8-1 gives various data about five different
thrust chambers with different kinds of propellants, cooling methods, injectors,
feed systems, thrust levels, or nozzle expansions. Some engine parameters are
also listed. Some of the terms used in this table will be explained later in this
chapter.

The basic analyses for thrust chamber specific impulse and combustion temper-
ature are given in Chapter 5, the basic design parameters (thrust, flow, chamber
pressure, or throat area) are in Chapter 3, and the combustion phenomena in
Chapter 9.

Although we use the word thrust chamber in this book (for rocket engines
generally larger than 1000 Ibf thrust), some technical publications use the terms
thrust cylinder , thrust cell, or rocket combustor . We will also use the term thruster
for small thrust units, such as attitude control thrusters, and for electrical propul-
sion systems.

271

272 THRUST CHAMBERS

FIGURE 8-1. Construction of an early regeneratively cooled tubular thrust chamber
using a kerosene-type fuel and liquid oxygen, as originally used in the Thor missile.
The nozzle throat inside diameter is about 15 in. The sea-level thrust was originally
120,000 Ibf, hut was uprated to 135,000, then 150,000, and finally to 165,000 Ibf by
increasing the flow and chamber pressure and strengthening and modifying the hardware.
The cone-shaped nozzle exit was replaced by a bell-shaped nozzle exit. Figure 8-9 shows
how the fuel flows down through every other tube and returns through the adjacent tube
before flowing into the injector. Figure 8-5 shows the flow passages in a similar injector.
(Courtesy of Rolls Royce, England.)

r — Stainless steel inner
\ liner with 120 groove
1 2 quarter-wave \ coolant passages and

273

MGUKli 8-2. Simplmed halt-section ot one ot the two thrust chambers ot the orbital maneuvering engines used on the Space Shuttle Orbiter.
Each develops a vacuum thrust of 6000 Ibf (26,689 N) and delivers a minimum vacuum specific impulse of 310 sec, using nitrogen tetroxide
and monomethyl hydrazine propellants at a nominal mixture ratio of 1.65 and a nominal chamber pressure of 128 psia. It is designed for
100 flight missions, a service life of 10 years, and a minimum of 500 starts. These engines provide the thrust for final orbit attainment, orbit
circularization, orbit transfer, rendezvous, and deorbit maneuvers. The nozzle area ratio is 55:1. (Courtesy of Aerojet Propulsion Company.)

TABLE 8-1. Thrust Chamber Characteristics

o c

CJ ^ -

O -Si, C -

O o

^ U lyj

^ II 5

c /3

“ifis

-Ills

H C bO

1-^ . . O

iS ^ ^ 3 o

u 3 ti 8 c

Q j 2 o -S

c/3

-o o
D h
•S C3
s o

2.0

1— ^

u ^
Mo g

s ^ K

x; c/3
c/3 p

« IS
2 K 3 3

o hM c/3

CQ u
.P ^

T3 '5

> >. B S

M 3-3 0

" O C 3 pi.

■d M .-3 ^ 3

S 3-3.3

3 > d U

K tH ^ F “
n o o>) d

ii g-^ 3 a

3 3-300

« = 2 ^ o
D CL, o c
QC -3

IT)

^ o
<C >o
Z

'sO

O o ^
p) g ^ ^
^ VO

c/3

»n

r-"

O 00
r-. 'so

’3

o^ §

»n

’3

c/3

^ <
^ Z

»o

^ g'ct S

.2 g.

2 M

D O

3 S
><

_ c? ct
.2-5 I

O 2 M

b 3 s

D Cfl >
c/3 "

D ^
Dh CJ

c/3 1)

CJ -C

(U X

c/3 1)

■g

1-H

c/3

1 .

) 1

■■^ 's

c/3 ^

^ ^
(U u — !

ti Q. ^

Oh >

i- C

3 -2 A
■§3 3
I c “

3 6J0 tH

x: (U

3 t2 -3

■2 -o i

cb c

:b c/3

U H

274

Gimbal mount sea-level dry weight (Ibf) < 10 57.3 NA 70 23

Chamber diameter (in.) — 15.75 1.09 21 11.7

Nozzle throat diameter (in.) 5.2 9.25 0.427 16.2 7.5

Nozzle exit diameter (in.) 88 68.28 3.018 45.8 60

Nozzle exit area ratio 285 54:1 50:1 8:1 65:1

IT) r-;
O 00

CO —

D O

»o

CL>

o ^

’Qh ^

*0 ^
2 •§
j= g
•a

o o o o S

g -t ^ g g

T3 ^

V5 cd

Oh ^

00 ^ ^
'sO CO 00

’o

o uo
> o

J=>

't-i cd
T3

CnJ o
o^ o

CN ^

-o

tin

O »/l o

^ CO
'O — in
in ^

Q- S

E S

s ^

^ 00

i> 00 in
O ■^* CO

^ .S
o

o 2
o ‘P

o < =

Z A ^

00 o
00

cN in

^ 't;

o '

o o
»n in

CS (N
»n in

== s

V3 Cd
V5 ■*-'

^ X3

cu ^

Q-

e 2

-- O

^ X
I CO X

3 A ^

:§ ^ <
r o

< »n

-fa

g s

o >
U S

2 ^ X X
^ in ^
CS (N

> CJ
^ >>
Q- «-•

P Vh
D- r-

o 2

X Q,
D &

e 2 “

•s J3
g M
3 g

(U 35

^ g
C -S
° S

cd cd
X X

U U

00 o
c >

1 — I 73

^ 's

o a

g ■§
•-§ §
! S •<-'
. Cd zJ

1 CJ

t .;_, cd

S "a

H Ph

D
"O
00
C .'

c« X

c ^

^ -
z u

X ^

5 «

3^ ^

^ 00
■p c

cd X
tc O

(3 O

e 2

X ^

! 3 o

, “ ^

) o ?o

S-(

; fa- i-i

^ B

: a;

• cd c

O o

ft o-.S’

2 S i-
-d -3 “
D X

fa 3

(U CJ

V- l-

== =3 ^

M IZ2

M C/3 Cj

a a ‘S

ft ft °

^ fa

2 o “

Engine thrust (at sea level) (lb) NA 190,400 12 165,000 NA

Engine thrust (altitude) (lb) 24,750 242,500 18 207,700 9850

Engine specific impulse at sea level NA 349.9 200 253 320

Engine specific impulse at altitude 462 445.6 290 288 320

Engine mixture ratio (oxidizer/fuel) 5.88 6.0 2.0 2.27 1.90

276 THRUST CHAMBERS

8.1. INJECTORS

The functions of an injector are similar to those of a carburetor of an internal
combustion engine. The injector has to introduce and meter the flow of liquid
propellants to the combustion chamber, cause the liquids to be broken up into
small droplets (a process called atomization), and distribute and mix the propel-
lants in such a manner that a correctly proportioned mixture of fuel and oxidizer
will result, with uniform propellant mass flow and composition over the chamber
cross section.

There are two design approaches for admitting propellants into the combustion
chamber. The individual locations on the injector face are called injector elements.
The older types use a set of propellant jets going through a set of holes in a pattern
on the injector face. Different hole arrangements are shown in Figure 8-3. Many
of the rocket injectors developed in the United States used this type for large
and small thrust chambers. The second type has cylindrical injection elements,
which are inserted and fastened (welded, brazed, or soldered) into the injector
face and each element delivers a conically shaped spray of propellants into the
combustion chamber. Figure 8-4 shows several spray injection elements with
conical sheets of propellant coming from a slot or from the internal edge of a
hollow cylinder in the injection element. It has been used with LOX/LH 2 thrust
chamber worldwide including the Space Shuttle thrust chamber. Also it has been
the preferred approach in Russia (and its predecessor the Soviet Union), where
it was used with most propellants and all thrust chamber sizes. There are also
some combination types that use jets and sprays together. Typical injectors with
hole patterns can be seen in Figs. 8-1 and 8-5.

The injection pattern on the face of the injector is closely related to the
internal manifolds or feed passages within the injector. These provide for the
distribution of the propellant from the injector inlet to all the injection holes or
spray elements. A large complex manifold volume allows low passage velocities
and good distribution of flow over the cross section of the chamber. A small
manifold volume allows for a lighter weight injector, a faster start, and reduces
the amount of “dribble” flow after the main valves are shut. The higher passage
velocities usually cause a more uneven flow through different identical injection
holes and thus a poorer distribution and wider local gas composition variation.
Dribbling results in after-burning, which is an inefficient irregular combustion
that gives a little “cutoff’ thrust after valve closing. For applications with very
accurate terminal vehicle velocity requirements, the cutoff impulse has to be very
small and reproducible and often valves are built into the injector to minimize
passage volume.

Doublet impinging-stream-type, multiple-hole injectors are commonly used
with oxygen-hydrocarbon and storable propellants, as seen in Fig. 8-3. For
unlike doublet patterns the propellants are injected through a number of separate
small holes in such a manner that the fuel and oxidizer streams impinge upon
each other. Impingement forms thin liquid fans and aids atomization of the liquids
into droplets, also aiding distribution. Characteristics of specific injector orifices

8.1. INJECTORS 277

Injection holes

Impingment

points

Doublet impinging
stream pattern

Fuel

manifolds

Impingment
point

Face of
injector

Triplet impinging
stream pattern

Typical impingement
point

Seif- impinging
stream pattern

Typical straight
propellant stream

Fuel
manifold^N^

Oxidizer

manifold

^Premix chamber

Injector lace

' Oxidizer inlet
^Fuel inlet

.Conical splash
plate

Splash plate
pattern

Premixing type

FIGURE 8-3. The upper four simplified sketches show common injection designs using
holes. The premix chamber-type (its igniter is not shown) is used today on a few large
LOX/LH 2 thrust chambers. The splash plate-type was popular in the late 1940s and 1950s
but is not used today. (Used with permission from Ref. 8-1.)

are given in Table 8-2. Impinging hole injectors are also used for like-on-like or
self-impinging patterns (fuel-on-fuel and oxidizer-on-oxidizer). The two liquid
streams then form a fan whieh breaks up into droplets. Unlike doublets work
best when the hole size (more exactly, the volume flow) of the fuel is about
equal to that of the oxidizer and the ignition delay is long enough to allow the
formation of fans. For uneven volume flow the triplet pattern often seems to be
more effective.

The nonimpinging or shower head injector employs nonimpinging streams
of propellant usually emerging normal to the face of the injector. It relies on
turbulence and diffusion to achieve mixing. The German World War II V-2
rocket used this type of injector. This type is no longer used because it requires
a large chamber volume for good combustion.

278 THRUST CHAMBERS

Injeaor face

(a) Single propellent spray element
%vitn spiral Insert.

Tangential '
Injeaion v
orifices \\
(to make \\
flow swirl) V

Iq. I’”

(cf) Coaxial blpropellant

injection spray element with
external burning.

. Conical
X spray

Injeaor face

(b)Slngle propellant spray element
with tangential injection.

Oxidizer , _ ,

Manifold^ I

Tangential
injection ^
orifices \\
(to make \\
flow swirl) V

^ impingment
outside

'' Injector
(e) Coaxial blpropellant face
injection sprayelement
with Internal burning.

-Cone spray
without
prerotation
will break up
Into small
droplets

Dual valve
seats for
shutoff

Injector face' clings to
(0 Coaxial bipropellant
pintle element with
movable pintle (throttling)
and dual-face valves (for shutoff).

FIGURE 8-4. Simplified cross-section sketches of several common types of spray injec-
tion elements. (Used with permission from Ref. 8-1.)

Sheet or spray-type injectors give cylindrical, conical, or other types of spray
sheets; these sprays generally intersect and thereby promote mixing and atom-
ization. See Fig. 8-4. The conical sheets break up into droplets, which are then
vaporized. The size distribution of droplets from spray injection elements is usu-
ally more uniform than with impinging streams from holes. By changing some
of the internal dimensions of the spray elements (change size or number of tan-
gential feed holes, the length/protrusion of an internal cylinder, or the angle of
an internal spiral rib), it is possible to change the angle of the conical sheet or
the location of the impingement of fuel and oxidizer spray sheets, and to affect
the combustion efficiency. By varying the width of the sheet (through an axi-
ally movable sleeve), it is possible to throttle the propellant flow over a wide
range without excessive reduction in injector pressure drop. This type of variable
area concentric tube injector was used on the descent engine of the Apollo lunar
excursion module and throttled over a 10:1 range of flow with only a very small
change in mixture ratio or performance.

8.1. INJECTORS 279

FIGURE 8-5. Injector with 90° self-impinging (fuel-against-fuel and oxidizer-against-
oxidizer)-type countersunk doublet injection pattern. Large holes are inlets to fuel mani-
folds. Predrilled rings are brazed alternately over an annular fuel manifold or groove and
a similar adjacent oxidizer manifold or groove. A section through a similar but larger
injector is shown in Fig. 8-1.

The coaxial hollow post or spray injector has heen used for liquid oxygen
and gaseous hydrogen injectors hy most domestic and foreign rocket designers.
It is shown in sketches d and e of Fig. 8-4. It works well when the liquid
hydrogen has absorbed heat from cooling jackets and has been gasified. This
gasified hydrogen flows at high speed (typically 330 m/sec or 1000 ft/sec); the
liquid oxygen flows far more slowly (usually at less than 33 m/sec or 100 ft/sec),
and the differential velocity causes a shear action, which helps to break up the
oxygen stream into small droplets. The injector has a multiplicity of these coaxial
elements on its face. In Russia and in Germany the spray injector elements have
also been used with storable propellants.

The SSME injector shown in Fig. 9-6 uses 600 of these concentric sleeve
injection elements; 75 of them have been lengthened beyond the injector face to
form cooled baffles, which reduce the incidence of combustion instability.

The original method of making injection holes was to carefully drill them and
round out or chamfer their inlets. This is still being done today. It is difficult
to align these holes accurately (for good impingement) and to avoid burrs and
surface irregularities. One method that avoids these problems and allows a large
number of small accurate injecton orifices is to use multiple etched, very thin
plates (often called platelets) that are then stacked and diffusion bonded together

280 THRUST CHAMBERS

TABLE 8-2. Injector Discharge Coefficients

Orifice Type

Discharge

Diagram Diameter (mm) Coefficient

Sharp-edged orifice

Above 2.5 0.61

Below 2.5 0.65 approx.

Short-tube with rounded
entrance L/D >3.0

1.00 0.88

1.57 0.90

1.00

(with L/D ~ 1.0 ) 0.70

Short tube with conical
entrance

0.50 0.7

1.00 0.82

1.57 0.76

2.54 0.84-0.80

3.18 0.84-0.78

Short tube with spiral effect

1.0-6.4 0.2-0.55

Sharp-edged cone

1.00 0.70-0.69

1.57 0.72

to form a monolithic structure as shown in Fig. 8-6. The photo-etched pattern
on each of the individual plates or metal sheets then provides not only for many
small injection orifices at the injector face but also for internal distribution or
flow passages in the injector and sometimes also for a fine-mesh filter inside the
injector body. The platelets can be stacked parallel to or normal to the injector
face. The finished injector has been called the platelet injector and has been
patented by the Aerojet Propulsion Company.

Injector Flow Characteristics

The differences of the various injector element configurations shown in Figs. 8-3
and 8-4 reflect themselves in different hydraulic flow-pressure relationships, dif-
ferent starting characteristics, atomization, resistance to self-induced vibrations,
and combustion efficiency.

The hydraulic injector characteristics can be evaluated accurately and can
be designed for orifices with the desired injection pressures, injection velocities,
flows, and mixture ratio. For a given thrust F and a given effective exhaust

8.1. INJECTORS 281

Face plate with
inclined injection holes
(laser drilled)

Fuel distribution
plate with etched
cooling passages

Oxidizer distribution
manifold plate

Alignment pins (2)

Cover plate with
oxidizer inlet

Four stacked plates
(with exaggerated thickness)

Oxidizer filter
Fuel manifold

Oxidizer manifold

Oxidizer
inlet port

Intergrat

Like on like
impinging orifices

Cross section through an etched plate

(h)

Enlarged detail segment of
one design of like-on-like impinging
orifices and feed passage geometry

FIGURE 8-6. Simplified diagrams of two types of injector using a bonded platelet
construction technique: (a) injector for low thrust with four impinging unlike doublet
liquid streams; the individual plates are parallel to the injector face; (b) like-on-like
impinging stream injector with 144 orifices; plates are perpendicular to the injector face.
(Courtesy of Aerojet Propulsion Company).

velocity c, the total propellant mass flow m is given by m = F fc from Eq. 2-6.
The relations between the mixture ratio, the oxidizer, and the fuel flow rates
are given by Eqs. 6-1 to 6-4. For the of an incompressible fluid through
hydraulic orifices,

Q = CdA^l^pjp

( 8 - 1 )

282 THRUST CHAMBERS

m = Qp = CdA^/lfTKp (8-2)

where Q is the volume flow rate, Q the dimensionless discharge coefficient,
p the propellant mass density, A the cross-sectional area of the orifice, and Ap
the pressure drop. These relationships are general and can be applied to any one
section of the propellant feed system, to the injector, or to the overall liquid flow
system.

For any given pressure drops the injection orifices usually determine the mix-
ture ratio and the propellant flows of the rocket unit. From Eqs. 6-1 and 8-2
the mixture ratio r is

r = tho/mf = [iCd)o/{Cd)f]iAo/Af)y/ipo/pf){Apo/Apf) (8-3)

The quantities in the preceding equations have to be chosen so that the correct
design mixture ratio is attained. Orifices whose discharge coefficients are constant
over a large range of Reynolds numbers and whose ratio {Cd)o/{Cd)f remains
essentially invariant should be selected. For a given injector it is usually difficult
to maintain the mixture ratio constant at low flows or thrusts, such as in starting.

The quality of the injector can be checked by performing cold tests with
inert simulant liquids instead of reactive propellant liquids. Often water is used
to confirm pressure drops through the fuel or oxidizer side at different flows,
and this allows determination of the pressure drops with propellants and the
discharge coefficients. Nonmixable inert liquids are used with a special apparatus
to determine the local cold flow mixture ratio distribution over the chamber cross
section. The simulant liquid should be of approximately the same density and
viscosity as the actual propellant. New injectors are usually hot fired and tested
with actual propellants.

The actual mixture ratio can be estimated from cold flow test data, the mea-
sured hole areas, and discharge coefficients by correcting by the square root of the
density ratio of the simulant liquid and the propellant. When water at the same
pressure is fed alternately into both the fuel and the oxidizer sides, Apf = Ap„
and pf = po and the water mixture ratio will be

r = [{Cd)o/{Cd)f]AJAf (8-4)

Therefore, the mixture ratio measured in water tests can be converted into the
actual propellant mixture ratio by multiplying it by the square root of the density
ratio of the propellant combination and the square root of the pressure drop ratio.
The mechanism of propellant atomization with simultaneous vaporization, partial
combustion, and mixing is difficult to analyze, and performance of injectors has to
be evaluated by experiment within a burning rocket thrust chamber. The injection
velocity is given by

V = Q/A = CdyjlApjp

(8-5)

8.1. INJECTORS 283

Values of discharge coefficients for various types of injection orifices are shown
in Table 8-2. The velocity is a maximum for a given injection pressure drop when
the discharge coefficient approaches 1 . Smooth and well-rounded entrances to the
injection holes and clean bores give high values of the discharge coefficient, and
this hole entry design is the most common. Small differences in chamfers, hole
entry radius, or burrs at the edge of a hole can cause significant variations in
the discharge coefficient and the jet flow patterns, and these in turn can alter the
quality and distribution of the atomized small droplets, the local mixture ratio,
and the local heat transfer rates. An improperly manufactured hole can cause
local chamber or injector burnout.

When an oxidizer and a fuel jet impinge, the resultant momentum can be
calculated from the following relation, based on the principle of conservation of
momentum. Figure 8-7 illustrates a pair of impinging jets and defines j/q the
angle between the chamber axis and the oxidizer stream, y/ as the angle between
the chamber axis and the fuel stream, and 8 as the angle between the chamber
axis and the average resultant stream. If the total momentum of the two jets
before and after impingement is equal.

tan 8 =

nioVo sin /o — tn/Vf sin yy
thoVo cos Yo + thfVf cos yy

( 8 - 6 )

Good performance is often obtained when the resultant momentum of impinging
streams is approximately axial. If the resultant momentum is along the chamber
axis, 8 = 0, tan 5 = 0, and the angular relation for an axially directed jet momen-
tum is given by

mono sin yo = rhfVf sin yy (8-7)

From these equations the relation between yf,yo, and 8 can be determined. A
sample injector analysis is shown in Section 8.8.

Line of resultant
jet momentum

FIGURE 8-7. Angular relation of doublet impinging-stream injection pattern.

284 THRUST CHAMBERS

Factors Influencing Injector Behavior

A complete theory relating injector design parameters to rocket performance and
combustion phenomena has not yet been devised, and therefore the approach to
the design and development of liquid propellant rocket injectors has been largely
empirical. Some of the analyses (see Ref. 8-2) has been useful in understanding
the phenomenology and in indicating the directions for injector development. The
available data indicate several important factors that affect the performance and
operating characteristics of injectors; some of these are briefly enumerated here.
These factors are different for injectors where both propellants are liquid (such
as nitrogen tetroxide and hydrazine) or where one propellant is liquid and the
other is gaseous as with LOX/gasified liquid hydrogen. Monopropellant injectors
feed propellants into a catalyst bed and they are different and described in
Section 8.3.

Propellant Combination. The particular combination of fuel and oxidizer
affects such characteristics as the relative chemical reactivity, the ease and speed
of vaporization, the ignition temperature, the diffusion of hot gases, the volatil-
ity, or the surface tension. Hypergolic (self-igniting) propellants generally require
injector designs somewhat different from those required by propellants that must
be ignited. Injector designs that perform efficiently with one combination do not
necessarily work too well with a different propellant combination.

Injection, Element Pattern, and Orifice Size. With individual elements of
holes or sprays in the injector plate, there appears to be an optimum performance
and/or heat transfer condition for each of the following parameters: orifice size,
angle of impingement, angle of resultant momentum, distance of the impingement
locus from the injector face, number of injection orifices per unit of injector face
surface, flow per unit of injection orifice, and distribution of orifices over the
injector face. These parameters are largely determined experimentally or from
similar earlier successful injectors.

Transient Conditions. Starting and stopping may require special provisions
(temporary plugging of holes, accurate valve timing, insertion of paper cups over
holes to prevent entry of one propellant into the manifold of the other propellant
as was done on the German A-4 or V-2 thrust chamber, or check valves) to
permit satisfactory transient operation.

Hydraulic Characteristics. The orifice type and the pressure drop across the
injection orifice determine the injection velocity. A low pressure drop is desirable
to minimize the weight of the feed system or the pumping power. High-pressure
drops are used often to increase the rocket’s resistance to combustion instability
and enhance atomization of the liquids and improve performance.

8.2. COMBUSTION CHAMBER AND NOZZLE 285

Heat Transfer. Injectors influence the performance and the heat transferred
in rocket thrust chambers. Low heat transfer rates have been obtained when
the injection pattern resulted in an intentionally rich mixture near the chamber
walls and when the chamber pressure is low. In general, the higher performance
injectors have a higher heat transfer rate to the walls of the combustion chamber,
the nozzle, and the injector face.

Structural Design. The injector is highly loaded by pressure forces from the
combustion chamber and the propellant manifolds. During transition (starting or
stopping) these pressure conditions can cause internal transient structural stresses
which sometimes exceed the steady-state operating conditions. The faces of most
modem injectors are flat and must be reinforced by suitable structures which
nevertheless should provide no obstmctions to the hydraulic manifold passages;
the structure must also be sufficiently flexible to allow thermal deformations
caused by heating the injector face with hot combustion gases or cooling of
certain flow passages by cryogenic propellants. The injector design must also
provide for positive seals between fuel and oxidizer manifolds (an internal leak
can cause manifold explosions or internal fires) and a sealed attachment of the
injector to the chamber. In large, gimbal-mounted thmst chambers the injector
also often carries the main thrust load, and a gimbal mount is often directly
attached to the injector, as shown in Figs. 6-1 and 8-1.

Combustion Stability. The injection hole or spray pattern, impingement pat-
tern, hole or spray element distribution, and pressure drop have a strong influence
on combustion stability; some types are much more resistant to pressure distur-
bances. As explained in Section 9.3, the resistance to vibration is determined
experimentally, and often special antivibration devices, such as baffles or reso-
nance cavities, are designed directly into the injeetor.

8.2. COMBUSTION CHAMBER AND NOZZLE

The combustion chamber is that part of a thrust chamber where the eombustion
or burning of the propellant takes place. The combustion temperature is much
higher than the melting points of most chamber wall materials. Therefore it is
necessary either to cool these walls (as described in a later section of this chapter)
or to stop rocket operation before the critical wall areas become too hot. If the
heat transfer is too high and thus the wall temperatures become locally too high,
the thrust chamber will fail. Heat transfer to thrust chambers will be described
later in this chapter. Section 8.8 gives a sample analysis of a thrust chamber and
Ref. 8-2 describes the analyses for design and development.

Volume and Shape

Spherical chambers give the least internal surface area and inert mass per unit
chamber volume; they are expensive to build and several have been tried. Today

286 THRUST CHAMBERS

we prefer a cylindrical chamber (or slightly tapered cone frustum) with a flat
injector and a converging -diverging nozzle. The chamber volume is defined as
the volume up to the nozzle throat section, and it includes the cylindrical chamber
and the converging cone frustum of the nozzle. Neglecting the effect of the corner
radii, the chamber volume Vc is

y, = AiLi +AiL,(l + VA,/Ai +A,Mi) (8-8)

Here Li is the cylinder length, At/Ai is the chamber contraction ratio, and Lc is
the length of the conical frustum. The chamber surfaces exposed to heat transfer
from hot gas are the injector face, the inner surface of the cylinder chamber, and
the inner surface of the converging cone frustrum. The volume and shape are
selected after evaluating these parameters:

1 . The volume has to be large enough for adequate atomization, mixing, evapo-
ration, and complete combustion of propellants. Chamber volumes vary for
different propellants with the time delay necessary to vaporize and activate
the propellants and with the speed of combustion reaction of the propel-
lant combination. When the chamber volume is too small, combustion is
incomplete and the performance is poor. With higher chamber pressures
or with highly reactive propellants, and with injectors that give improved
mixing, a smaller chamber volume is usually permissible.

2. The chamber diameter and volume can influence the cooling requirements .
If the chamber volume and the chamber diameter are large, the heat transfer
rates to the walls will be reduced, the area exposed to heat will be large, and
the walls are somewhat thicker. Conversely, if the volume and cross section
are small, the inner wall surface area and the inert mass will be smaller, but
the chamber gas velocities and the heat transfer rates will be increased. For
the same thrust chamber there is an optimum chamber volume and diameter
where the total heat absorbed by the walls will be a minimum. This is
important when the available cooling capacity of the coolant is limited (e.g.,
oxygen-hydrocarbon at high mixture ratios) or if the maximum permissive
coolant temperature has to be limited (for safety reasons with hydrazine
cooling). The total heat transfer can also be further reduced by going to a
rich mixture ratio or by adding film cooling (discussed below).

3. All inert components should have minimum mass. The thrust chamber mass
is a function of the chamber dimensions, chamber pressure, and nozzle area
ratio, and the method of cooling.

4. Manufacturing considerations favor a simple chamber geometry, such as
a cylinder with a double cone bow-tie-shaped nozzle, low-cost materials,
and simple fabrication processes.

5. In some applications the length of the chamber and the nozzle relate directly
to the overall length of the vehicle. A large-diameter but short chamber
and/or a short nozzle can allow a somewhat shorter vehicle with a lower
structural inert vehicle mass.

8.2. COMBUSTION CHAMBER AND NOZZLE 287

6. The gas pressure drop for accelerating the combustion products within the
chamber should be a minimum; any pressure reduction at the nozzle inlet
reduces the exhaust velocity and the performance of the vehicle. These
losses become appreciable when the chamber area is less than three times
the throat area.

7. For the same thrust the combustion volume and the nozzle throat area
become smaller as the operating chamber pressure is increased. This means
that the chamber length and the nozzle length (for the same area ratio) also
decrease with increasing chamber pressure. The performance and the heat
transfer also go up with chamber pressure.

The preceding chamber considerations conflict with each other. It is, for
instance, impossible to have a large chamber that gives complete combustion
but has a low mass. Depending on the application, a compromise solution that
will satisfy the majority of these considerations is therefore usually selected and
verified by experiment.

The characteristic chamber length is defined as the length that a chamber of
the same volume would have if it were a straight tube and had no converging
nozzle section:

L* = Vc/A, (8-9)

where L* (pronounced el star) is the characteristic chamber length. At is the
nozzle throat area, and Vc is the chamber volume. The chamber includes all the
volume up to the throat area. Typical values for L* are between 0.8 and 3.0 m (2.6
to 10 ft) for several bipropellants and higher for some monopropellants. Because
this parameter does not consider any variables except the throat area, it is useful
only for a particular propellant combination and a narrow range of mixture ratio
and chamber pressure. The parameter L* was used about 40 years ago, but today
the chamber volume and shape are chosen by using data from successful thrust
chambers of prior similar designs and identical propellants.

The stay time t, of the propellant gases is the average value of the time spent
by each molecule or atom within the chamber volume. It is defined by

f, = Vc/{mVi) (8-10)

where m is the propellant mass flow, Vi is the average specific volume or volume
per unit mass of propellant gases in the chamber, and Vc is the chamber vol-
ume. The minimum stay time at which a good performance is attained defines the
chamber volume that gives essentially complete combustion. The stay time varies
for different propellants and has to be experimentally determined. It includes the
time necessary for vaporization, activation, and complete burning of the propel-
lant. Stay times have values of 0.001 to 0.040 sec for different types and sizes
of thrust chambers and for different propellants.

288 THRUST CHAMBERS

The nozzle dimensions and configuration can be determined from the analyses
presented in Chapter 3. The converging section of the nozzle experiences a much
higher internal gas pressure than the diverging section and therefore the design
of the converging wall is similar to the design of the cylindrical chamber wall.
The exact contour of the converging nozzle section is not critical. Most thrust
chambers use a shortened bell shape for the diverging nozzle section. Nozzles
with area ratios over 400 have been developed and flown.

In Chapter 3 it was stated that very large nozzle exit area ratios allow a
small but significant improvement in specific impulse, particularly at very high
altitudes; however, the extra length, extra nozzle mass, and extra vehicle mass
necessary to house a large nozzle often make this unattractive. This disadvantage
can be mitigated by a multipiece nozzle that is stored in annular pieces around
the engine during the ascent of the launch vehicle and automatically assem-
bled in space after launch vehicle separation and before firing. This concept,
known as extendible nozzle, has been successfully employed in solid propellant
rocket motors for space applications for about 27 years. The first flight with an
extendible nozzle on a liquid propellant engine was performed in 1998 with a
modified version of a Pratt & Whitney Rocketdyne upper stage engine. Its flight
performance is listed in Table 8-1. The engine is shown later in Fig. 8-17 and
its carbon-carbon extendible nozzle cone is described in the section on materials
and fabrication of this chapter.

Heat Transfer Distribution

Heat is transmitted to all internal hardware surfaces exposed to hot gases, namely
the injector face, the internal chamber and nozzle walls. The heat transfer rate
or heat transfer intensity, that is, local wall temperatures and heat transfer per
unit wall area, varies within the thrust chamber. A typical heat transfer rate
distribution is shown in Fig. 8-8. Only 0.5 to 5% of the total energy generated
in the gas is transferred as heat to the chamber walls. For a typical rocket of
44,820 N or 10,000 Ibf thrust the heat rejection rate to the wall may be between
0.75 and 3.5 MW, depending on the exact conditions and design. See Section 8.5.

The amount of heat transferred by conduction from the chamber gas to the
walls in a rocket thrust chamber is negligible. By far the largest part of the heat is
transferred by means of convection. A part (usually 5 to 35%) of the transferred
heat is attributable to radiation.

For constant chamber pressure, the chamber wall surface increases less rapidly
than the volume as the thrust level is raised. Thus the cooling of chambers is
generally easier in large thrust sizes, and the capacity of the wall material or the
coolant to absorb all the heat rejected by the hot gas is generally more critical
in smaller sizes because of the volume -surface relationship.

Higher chamber pressure leads to higher vehicle performance (higher /j), but
also to higher engine inert mass. However, the resulting increase of heat transfer
with chamber pressure often imposes design or material limits on the maximum
practical chamber pressure for both liquid and solid propellant rockets.

8.2. COMBUSTION CHAMBER AND NOZZLE

289

FIGURE 8-8. Typical axial heat transfer rate distribution for liquid propellant thrust
chambers and solid propellant rocket motors. The peak is always at the nozzle throat or
slightly upstream of the throat and the lowest value is usually near the nozzle exit.

The heat transfer intensity in chemical rocket propulsion can vary from less
than 50 W/cm^ or 0.3 Btu/in.^-sec to over 16 kW/cm^ or 100 Btu/in.^-sec. The
high values are for the nozzle throat region of large bipropellant thrust chambers
and high-pressure solid rocket motors. The lower values are for gas generators,
nozzle exit sections, or small thrust chambers at low chamber pressures.

Cooling of Thrust Chambers

The primary objective of cooling is to prevent the chamber and nozzle walls
from becoming too hot, so they will no longer be able to withstand the imposed
loads or stresses, thus causing the chamber or nozzle to fail. Most wall materials
lose strength and become weaker as temperature is increased. These loads and
stresses are discussed in the next section. With further heating, the walls would
ultimately fail or even melt. Cooling thus reduces the wall temperatures to an
acceptable value.

Basically, there are two cooling methods in common use today. The first is
the steady-state method. The heat transfer rate and the temperatures of the cham-
bers reach thermal equilibrium. This includes regenerative cooling and radiation
cooling. The duration is limited only by the available supply of propellant.

Regenerative cooling is done by building a cooling jacket around the thrust
chamber and circulating one of the liquid propellants (usually the fuel) through
it before it is fed to the injector. This cooling technique is used primarily with
bipropellant chambers of medium to large thrust. It has been effective in applica-
tions with high chamber pressure and high heat transfer rates. Also, most injectors
use regenerative cooling on their hot faces.

In radiation cooling the chamber and/or nozzle have only a single wall made
of high-temperature material, such as niobium or molybdenum. When it reaches

290 THRUST CHAMBERS

thermal equilibrium, this wall usually glows red or white hot and radiates heat
away to the surroundings or to empty space. Radiation cooling is used with
bipropellant and monopropellant thrusters, bipropellant and monopropellant gas
generators, and for diverging nozzle exhaust sections beyond an area ratio of
about 6 to 10 in larger thrust chambers (see Fig. 8-2). A few small bipropellant
thrusters are also radiation cooled. This cooling scheme has worked well with
lower chamber pressures (less than 250 psi) and moderate heat transfer rates.

The second cooling method relies on transient heat transfer or unsteady heat
transfer. It is also called heat sink cooling. The thrust chamber does not reach a
thermal equilibrium, and temperatures continue to increase with operating dura-
tion. The heat-absorbing capacity of the metal hardware determines its maximum
duration, which is usually short (a few seconds for an all-metal construction).
The rocket combustion operation has to be stopped just before any of the exposed
walls reaches a critical temperature at which it could fail. This method has mostly
been used with low chamber pressures and low heat transfer rates. Heat sink cool-
ing of thrust chambers can also be done by absorbing heat in an inner liner made
of an ablative material, such as hber-reinforced plastics. Ablative materials burn
or erode slowly and cumulative operating durations can be minutes. Ablative
materials are used extensively in solid propellant rocket motors and will be dis-
cussed further in Chapter 15. The analysis of both of these cooling methods is
given in the next section of this chapter.

Film cooling and special insulation are supplementary techniques that are used
occasionally with both methods to locally augment their cooling capability. All
these cooling methods will be described further in this chapter.

Cooling also helps to reduce the oxidation of the wall material and the rate
at which walls would be eaten away. The rates of chemical oxidizing reactions
between the hot gas and the wall material can increase dramatically with wall
temperature. This oxidation problem can be minimized not only by limiting the
wall temperature but also by burning the liquid propellants at a mixture ratio
where the percentage of aggressive gases in the hot gas (such as oxygen or
hydroxyl) is very small, and by coating certain wall materials with an oxidation-
resistant coating; for example, iridium has been coated on the inside of rhenium
walls.

Cooling with Steady-State Heat Transfer. Cooled thrust chambers have
provisions for cooling some or all metal parts coming into contact with hot gases,
such as chamber walls, nozzle walls, and injector faces. Internal cooling passages,
cooling jackets, or cooling coils permit the circulation of a coolant. Jackets can
consist of separate inner and outer walls or of an assembly of contoured, adjacent
tubes (see Figs. 8-1 and 8-9). The inner wall confines the gases, and the spaces
between the walls serve as the coolant passage. The nozzle throat region is
usually the location that has the highest heat transfer intensity and is therefore
the most difficult to cool. For this reason the cooling jacket is often designed
so that the coolant velocity is highest at the critical regions by restricting the
coolant passage cross section, and so that the fresh cold coolant enters the jacket

8.2. COMBUSTION CHAMBER AND NOZZLE 291

Top view without
manifold

Exit (Section C)

ocnc

Chamber
(Section B)
Throat
(Section A)

FIGURE 8-9. Diagram of a tubular cooling jacket. The tubes are bent to the chamber
and nozzle contours; they are formed hydraulically to give a variable cross section to
permit the same number of tubes at the throat and exit diameters. Coolant enters through
the inlet manifold into every other tube and proceeds axially to the nozzle exit manifold,
where it then enters the alternate tubes and returns axially to go directly to the injector.

at or near the nozzle. While the selection of the coolant velocity and its variation
along the wall for any given thrust chamber design depends on heat transfer
considerations, the selection of the coolant passage geometry often depends on
pressure loss, stresses, and manufacturing considerations. An axial flow cooling
jacket, or a tubular wall, has a low hydraulic friction loss but is practical only
for large coolant flows (above approximately 9 kg/sec). For small coolant flows
and small thrust units, the design tolerances of the cooling jacket width between
the inner and outer walls or the diameters of the tubes, become too small, or
the tolerances become prohibitive. Therefore, most small thrust chambers use
radiation cooling or ablative materials.

In regenerative cooling the heat absorbed by the coolant is not wasted; it
augments the initial energy content of the propellant prior to injection, increasing
the exhaust velocity slightly (0.1 to 1.5%). This method is called regenerative
cooling because of the similarity to steam regenerators. The design of the tubular
chamber and nozzle combines the advantages of a thin wall (good for reducing
thermal stresses and high wall temperatures) and a cool, lightweight structure.
Tubes are formed to special shapes and contours (see Figs. 8-1 and 8-9), usually
by hydraulic means, and then brazed, welded, or soldered together (see Ref. 8-3).
In order to take the gas pressure loads in hoop tension, they are reinforced on
the outside by high-strength bands or wires. While Fig. 8-9 shows alternate
tubes for up and down flow, there are chambers where the fuel inlet manifold
is downstream of the nozzle throat area and where the coolant flow is up and
down in the nozzle exit region, but unidirectionally up in the throat and chamber
regions.

Radiation cooling is another steady-state method of cooling. It is simple and
is used extensively in the low heat transfer applications listed previously. Fur-
ther discussion of radiation cooling is given in Section 8.4. In order for heat to
be radiated into space, it is usually necessary for the bare nozzle and chamber

292 THRUST CHAMBERS

to Stick out of the vehicle. Figure 8-16 shows a radiation-cooled thrust cham-
ber. Since the white hot glowing radiation-cooled chambers and/or nozzles are
potent radiators, they may cause undesirable heating of adjacent vehicle or engine
components. Therefore, many have insulation (see Fig. 8-13) or simple exter-
nal radiation shields to minimize these thermal effects; however, in these cases
the actual chamber or nozzle wall temperatures are higher than they would be
without the insulation or shielding.

Cooling with Transient Heat Transfer. Thrust chambers with unsteady heat
transfer are basically of two types. One is a simple metal chamber (steel, copper,
stainless steel, etc.) made with walls sufficiently thick to absorb plenty of heat
energy. For short-duration testing of injectors, testing of new propellants, rating
combustion stability, and very short duration rocket-propelled missiles, such as
an antitank weapon, a heavy-walled simple, short-duration steel chamber is often
used. The common method of ablative cooling or heat sink cooling uses a combi-
nation of endothermic reactions (breakdown or distillation of matrix material into
smaller compounds and gases), pyrolysis of organic materials, counter-current
heat flow and coolant gas mass flow, charring, and localized melting. An abla-
tive material usually consists of a series of strong, oriented fibers (such as glass,
Kevlar, or carbon fibers) engulfed by a matrix of an organic binder material (such
as plastics, epoxy resins or phenolic resins). As shown in Fig. 15-11, the gases
seep out of the matrix and form a protective film cooling layer on the inner wall
surfaees. The fibers and the residues of the matrix form a hard char or porous
coke-like material that preserves the wall contour shapes.

The orientation, number, and type of fiber determine the ability of the com-
posite ablative material to withstand significant stresses in preferred directions.
For example, internal pressure produces longitudinal as well as hoop stresses in
the thrust chamber walls and thermal stresses produce compression on the inside
of the walls and tensile stresses on the outside. We have learned how to place the
fibers in two or three directions, which makes them anisotropic. We then speak
of two-dimensional (2-D) and 3-D fiber orientation.

A set of strong carbon fibers in a matrix of amorphous carbon is a special,
but favorite type of material. It is often abbreviated as C-C or carbon-carbon.
The carbon materials lose their ability to carry loads at a high temperature of
about 3700 K or 6200 F. Carbon oxidizes readily form CO or CO 2 . Its best
applications are with fuel-rich propellant mixtures that have little or no free
oxygen or hydroxyl in their exhaust. It has also been used for nozzle throat
inserts. Properties for one type of C-C are given in Table 15-5. AC-C nozzle
extension is shown in Fig. 8-17. See Ref. 8-4.

Ablative cooling was first used and is still used extensively with solid pro-
pellant rocket motors. It has since been successfully applied to liquid propel-
lant thrust chambers, particularly at low chamber pressure, where the static gas
temperatures are relatively low. It is still used today as a material for nozzle
extensions, such as for the RS-68 in Fig. 6-10, and it can operate for several
minutes. It is also used as a chamber and nozzle liner at low chamber pres-
sure. An example is the axial gimbaled thruster of the Peacekeeper fourth stage

8.2. COMBUSTION CHAMBER AND NOZZLE 293

which is seen in Figure 6-13. Ablative lined small thrusters (100 Ibf thrust or
less) have been flown extensively in the 1950s and 1960s in Apollo missions
and other applications for attitude control and minor maneuvers. They are no
longer used today because they are relatively heavy and because the eroded par-
ticle, droplets, or certain gases in the exhaust plume deposit condense on optical
surfaces of spacecraft (mirror, solar cells, or windows).

It is often advantageous to use a different cooling method for the downstream
part of the diverging nozzle section, because its heat transfer rate per unit area
is usually much lower than in the chamber or the converging nozzle section,
particularly with nozzles of large area ratio. There is usually a small saving in
inert engine mass, a small increase in performance, and a cost saving, if the
chamber and the converging nozzle section and the throat region (up to an area
ratio of perhaps 5 to 10) use regenerative cooling and the remainder of the nozzle
exit section is radiation cooled (or sometimes ablative cooled). See Fig. 8-2 and
Ref. 8-4.

Film Cooling. This is an auxiliary method applied to chambers and/or nozzles,
augmenting either a marginal steady-state or a transient cooling method. It can
be applied to a complete thrust chamber or just to the nozzle, where heat transfer
is the highest. Film cooling is a method of cooling whereby a relatively cool
thin fluid film covers and protects exposed wall surfaces from excessive heat
transfer. Figure 8-10 shows film-cooled chambers. The film can be introduced
by injecting small quantities of fuel or an inert fluid at very low velocity through
a large number of orifices along the exposed surfaces in such a manner that a
protective relatively cool gas film or cool boundary layer is formed. In liquid
propellant rocket engines extra fuel can also be admitted through extra injection
holes at the outer layers of the injector; thus a propellant mixture is achieved (at
the periphery of the chamber), which has a much lower combustion temperature.
This differs from film cooling or transpiration cooling because there does not
have to be a chamber cooling jacket or film-cooling manifolds.

Film cooling by itself (without another cooling method) has been very effective
in keeping chamber and nozzle materials from becoming too hot. In fact the very
first thmst chambers developed by Robert H. Goddard in the 1920s were film
cooled. However, the film coolant is not burned effectively and there was a 5
to 17% reduction in specific impulse. Therefore, today, film cooling is used in
small quantities (1 to 6% of fuel) to locally supplement other cooling methods
and the performance loss is 0.5 to 2%. In solid propellant rocket engines this
film cooling can be accomplished by inserting a ring of cool-burning propellant
upstream of the nozzle, as shown in Fig. 8-10 or by wall insulation materials,
whose ablation and charring will release relatively cool gases into the boundary
layer.

Turbine discharge gas (400 to 800°C or 752 to 1972°F) has also been used as
a film coolant for uncooled nozzle exit sections of large liquid propellant rocket
engines. Of course, the injection of an annular gas layer at the periphery of the
nozzle exit, at a temperature lower than the maximum possible value, causes

294 THRUST CHAMBERS

FIGURE 8-10. Simplified diagrams of three different methods of forming a cool bound-
ary layer in the nozzle.

a decrease in a specific impulse. Therefore, it is desirable to reduce both the
thickness of this cooler layer and the mass flow of cooler gas, relative to the
total flow, to a practical minimum value at which cooling will still be effective.

A special type of film cooling, sweat cooling or transpiration cooling, uses
a porous wall material which admits a coolant through pores uniformly over the
surface. This technique has been successfully used to cool injector faces in the
upper stage engine (J-2) of the moon launch vehicle and the Space Shuttle main
engine (SSME) with hydrogen fuel.

Thermal Insulation. Theoretically, a good thermal insulation layer on the gas
side of the chamber wall can be very effective in reducing chamber wall heat
transfer and wall temperatures. However, efforts with good insulation materials
such as refractory oxides or ceramic carbides have usually not been success-
ful. They will not withstand differential thermal expansion without cracking or
spalling. A sharp edge on the surface (crack or flaked-off piece of insulator)
will cause a sudden rise in the stagnation temperature and most likely lead to
a local wall failure. Asbestos is a good insulator and was used several decades
ago; because it is a cancer-causing agent, it is no longer used. Coating develop-
ment efforts with rhenium and other materials are continuing. Insulation or heat
shields have been successfully applied on the exterior of radiation-cooled thrust
chambers to reduce the heat transfer from the hot thrust chamber wall to adjacent
sensitive equipment or structures.

With hydrocarbon fuels it is possible to form small carbon particles or soot in
the hot gas and that can lead to a carbon deposit on the gas side of the chamber

8.2. COMBUSTION CHAMBER AND NOZZLE 295

or nozzle walls. If it is a thin, mildly adhesive soot, it can be an insulator, but it
is difficult to reproduce such a coating. More likely it forms hard, caked deposits,
which can be spalled off in localized flakes and form sharp edges, and then it is
undesirable. Most designers have preferred instead to use film cooling or extra
high coolant velocities in the cooling jacket with injectors that do not create
adhesive soot.

Hydraulic Losses in the Cooiing Passage

The cooling coil or cooling jacket should be designed so that the fluid adsorbs all
the heat transferred across the inner motor wall, and so that the coolant pressure
drop will be small.

A higher pressure drop allows a higher coolant velocity in the cooling jacket,
will do a better job of cooling, but requires a heavier feed system, which increases
the engine mass slightly and thus also the total inert vehicle mass. For many
liquid propellant rockets the coolant velocity in the chamber is approximately
3 to 10 m/sec or 10 to 33 ft/sec and, at the nozzle throat, 6 to 24 m/sec or 20 to
80 ft/sec.

A cooling passage can be considered to be a hydraulic pipe, and the friction
loss can be calculated accordingly. For a straight pipe,

Ap/p^\fvHL/D) ( 8 - 11 )

where Ap is the friction pressure loss, p the coolant mass density, L the length of
coolant passage, D the equivalent diameter, v the average velocity in the cooling
passage, and / the friction loss coefficient. In English engineering units the right
side of the equation has to be divided by go> the sea-level acceleration of gravity
(32.2 ft/sec^). The friction loss coefficient is a function of Reynolds number and
has values between 0.02 and 0.05. A typical pressure loss of a cooling jacket is
between 5 and 25% of the chamber pressure.

A large portion of the pressure drop in a cooling jacket usually occurs in those
locations where the flow direction or the flow-passage cross section is changed.
Here the sudden expansion or contraction causes a loss, sometimes larger than
the velocity head u^/2. This hydraulic situation exists in inlet and outlet chamber
manifolds, injector passages, valves, and expansion joints.

The pressure loss in the cooling passages of a thrust chamber can be calculated,
but more often it is measured. This pressure loss is usually determined in cold
flow tests (with an inert fluid instead of the propellant and without combustion),
and then the measured value is corrected for the actual propellant (different phys-
ical properties) and the hot firing conditions; a higher temperature will change
propellant densities or viscosities and in some designs it changes the cross section
of cooling flow passages.

Thrust Chamber Wall Loads and Stresses

The analysis of loads and stresses is performed on all propulsion components
during their engineering design. Its purpose is to assure the propulsion designer

296 THRUST CHAMBERS

and the flight vehicle user that (1) the components are strong enough to carry all
the loads, so that they can fulfill their intended function; (2) potential failures have
been identified, together with the possible remedies or redesigns; and (3) their
masses have been reduced to a practical minimum. In this section we concentrate
on the loads and stresses in the walls of thrust chambers, where high heat fluxes
and large thermal stresses complicate the stress analysis. Some of the information
on safety factors and stress analysis apply also to all propulsion systems, including
solid propellant motors and electric propulsion.

The safety factors (really margins for ignorance) are very small in rocket
propulsion when compared to commercial machinery, where these factors can
be 2 to 6 times larger. Several load conditions are considered for each rocket
component; they are:

1 . Maximum expected working load is the largest likely operating load under
all likely operating conditions or transients. Examples are operating at a
slightly higher chamber pressure than nominal as set by tolerances in design
or fabrication (an example is the tolerance in setting the tank pressure
regulator) or the likely transient overpressure from ignition shock.

2. The design limit load is typically 1.20 times the maximum expected work-
ing load, to provide a margin. If the variation in material composition,
material properties, the uncertainties in the method of stress analysis, or
predicted loads are significant, a larger factor may be selected.

3. The damaging load can be based on the yield load or the ultimate load
or the endurance limit load, whichever gives the lowest value. The yield
load causes a permanent set or deformation, and it is typically set as 1.10
times the design limit load. The endurance limit may be set by fatigue or
creep considerations (such as in pulsing). The ultimate load induces a stress
equal to the ultimate strength of the material, where significant elongation
and area reduction can lead to failure. Typically, it is set as 1.50 times the
design limit load.

4. The proof test load is applied to engines or their components during devel-
opment and manufacturing inspection. It is often equal to the design limit
load, provided this load condition can be simulated in a laboratory. Thrust
chambers and other components, whose high thermal stresses are difficult
to simulate, use actual hot firing tests to obtain this proof, often with loads
that approach the design limit load (e.g., higher than nominal chamber
pressure or a mixture ratio that results in hotter gas).

The walls of all thrust chambers are subjected to radial and axial loads from
the chamber pressure, flight accelerations (axial and transverse), vibration, and
thermal stresses. They also have to withstand a momentary ignition pressure
surge or shock, often due to excessive propellant accumulation in the chamber.
This surge can exceed the nominal chamber pressure. In addition, the chamber
walls have to transmit thrust loads as well as forces and in some applications
also moments, imposed by thrust vector control devices described in Chapter 18.

8.2. COMBUSTION CHAMBER AND NOZZLE 297

Walls also have to survive a “thermal shock,” namely, the initial thermal stresses
at rapid starting. When walls are cold or at ambient temperature, they experience
higher gas heating rates than after the walls have been heated. These loads are
different for almost every design, and each unit has to be considered individually
in determining the wall strengths.

A heat transfer analysis is usually done only for the most critical wall regions,
such as at and near the nozzle throat, at a critical location in the chamber, and
sometimes at the nozzle exit. The thermal stresses induced by the temperature
difference across the wall are often the most severe stresses and a change in
heat transfer or wall temperature distribution will affect the stresses in the wall.
Specific failure criteria (wall temperature limit, reaching yield stress, or maximum
coolant temperature, etc.) have to be established for these analyses.

The temperature differential across a chamber wall introduces a compressive
stress on the inside and a tensile stress on the outside of the inner wall; the stress
s can be calculated for simple cylindrical chamber walls that are thin in relation
to their radius as

s =2kE /S.T/{\-v) (8-12)

where X is the coefficient of thermal expansion of the wall material, E the
modulus of elasticity of the wall material, AT the temperature drop across the
wall, and v the Poisson ratio of the wall material. Temperature stresses frequently
exceed the yield point. The materials experience a change in the yield strength and
the modulus of elasticity with temperature. The preceding equation is applicable
only to elastic deformations. This yielding of rocket thrust chamber wall materials
can be observed by the small and gradual contraction of the throat diameter after
each operation (perhaps 0.05% reduction in throat diameter after each firing) and
the formation of progressive cracks of the inside wall surface of the chamber and
throat after successive runs. These phenomena limit the useful life and number
of starts or temperature cycles of a thrust chamber (see Refs. 8-5 and 8-6).

In selecting a working stress for the wall material of a thrust chamber, the
variation of strength with temperature and the temperature stresses over the wall
thickness have to be considered. The temperature drop across the inner wall is
typically between 50 and 550 K, and an average temperature is sometimes used
for estimating the material properties. The most severe thermal stresses can occur
during the start, when the hot gases cause thermal shock to the hardware. These
transient thermal gradients cause severe thermal strain and local yielding.

A picture of a typical steady-state stress distribution caused by pressure loads
and thermal gradients is shown in Fig. 8-11. Here the inner wall is subjected to a
compressive pressure differential caused by a high liquid pressure in the cooling
jacket and a relatively large temperature gradient. In a large rocket chamber,
such as is used in the Redstone missile or the German V-2, the wall thickness
of the nozzle steel way may be 7 mm and the temperature differential across it
may readily exceed several hundred degrees. This temperature gradient causes
the hot inner wall surface to expand more than the wall surface on the coolant

298 THRUST CHAMBERS

Gas film

Liquid

coolant

Liquid film

Typical temperature
distribution

Neutral axis (zero stress)

1. Stress due to thermal expansion
gradient across wall only

- 2. Stress due to pressure differential
across wall only

3. Resultant stress (sum of curves 1
and 2) with no yielding and constant
modulus of elasticity

■•4. Actual stress in wall with yielding
at hot gas side and changing
modulus of elasticity

5. Yield stress distribution across
wall (varies with temperature)

I^Wall thickness—

Distance from thrust chamber axis

FIGURE 8-11. Typical stresses in a thrust chamber inner wall.

side and imposes a high compressive thermal stress on the inside surface and
a high tensile thermal stress on the coolant side. In these thick walls the stress
induced by the pressure load is usually small compared to the thermal stress.
The resultant stress distribution in thick inner walls (shown shaded in the sample
stress diagram of Fig. 8-11) indicates that the stress in the third of the wall
adjacent to the hot gases has exceeded the yield point. Because the modulus of
elasticity and the yield point diminish with temperature, the stress distribution is
not linear over the yielded portion of the wall. In effect, this inner portion acts
as a heat shield for the outer portion which carries the load.

Because of the differential expansion between the hot inner shell and the
relatively cold outer shell, it is necessary to provide for axial expansion joints to
prevent severe temperature stresses. This is particularly critical in larger double-
walled thrust chambers. The German V-2 thrust chamber expanded over 5 mm
in an axial and 4 mm in a radial direction.

Tubes for tubular wall thrust chambers are subjected to several different
stress conditions. Only that portion of an individual cooling tube exposed to

8.2. COMBUSTION CHAMBER AND NOZZLE 299

hot chamber gases experiences high thermal stresses and deformation as shown
in Fig. 8-15. The tubes have to hold the internal coolant pressure, absorb the
thermal stresses, and contain the gas pressure in the chamber. The hottest tem-
perature occurs in the center of the outer surface of that portion of each tube
exposed to hot gas. The thermal stresses are relatively low, since the temperature
gradient is small; copper has a high conductivity and the walls are relatively thin
(0.5 to 2 mm). The coolant pressure-induced load on the tubes is relatively high,
particularly if the thrust chamber operates at high chamber pressure. The inter-
nal coolant pressure tends to separate the tubes. The gas pressure loads in the
chamber are usually taken by reinforcing bands which are put over the outside
of the tubular jacket assembly (see Figs. 8-1 and 8-9). The joints between the
tubes have to be gas tight and this can be accomplished by soldering, welding,
or brazing.

When a high-area-ratio nozzle is operated at high ambient pressure (sea level
or low altitude), the lower part of the nozzle structure experiences a compression
because the pressure in the nozzle near the exit is actually below atmospheric
value. Therefore, high-area-ratio nozzles usually have stiffening rings on the
outside of the nozzle near the exit to maintain a circular shape and thus prevent
buckling, flutter, or thrust misalignment.

Aerospike Thrust Chamber

A separate category comprises thrust chambers using a center body, such as a
plug nozzle or aerospike nozzle. They have more surface to cool than ordinary
thrust chambers. A circular aerospike thruster is described in Chapter 3 and
shown schematically in Figs. 3-12 and 8-12 and described in Ref. 8-7. Here the
diameter of the exhaust flow plume increases with altitude. Each individual small
(regeneratively cooled) thrust chamber or cell has its own cylindrical combustion
chamber, a circular small nozzle throat, but a rectangular nozzle exit of low area
ratio. The gas from these rectangular nozzle exits is further expanded (and thus
reaches a higher exhaust velocity) along the contour of the spike or ramp. The flat
surface at the bottom or base is porous or full of small holes and a low-pressure
gas flows through these openings. This causes a back pressure on the flat base
surface. This flow can be the exhaust gas from the turbopumps of an engine
with a gas generator cycle; it is typically 1 or 2% of the total flow. The gas
region below this base is enclosed by gas flows from the ramps. A linear version
of a truncated (shortened) aerospike engine with two opposing ramps has been
developed using LOX/LH 2 propellants at a thrust level of about 200,000 Ibf and
the ground tests were satisfactory. It has two fuel-cooled side plates. Except for
a flight of a very small, inefficient, truncated experimental aerospike, none of
these liquid propellant engines have flown.

The thrust of this aerospike thrust chamber consists of (1) the axial component
thrusts of each of the little chamber modules or cells, (2) the integral of the
pressures acting on the ramps over the axially projected area normal to the axis
of the ramps, and (3) the pressure acting over the base area.

300 THRUST CHAMBERS

Propellant
enters through

FIGURE 8-12. Pressure profile and flow pattern along the ramp of an aerospike nozzle.

The local gas pressures on the ramps are influenced by shock wave phenomena
and change pressure profiles with altitude. Figure 8-12 shows a typical pressure
distribution on a typical ramp surface and the flow patterns for low and high
altitude. The hot gas flows coming out of the cell nozzles are turned into a nearly
axial direction by multiple expansion waves (shown as dashed lines), which
cause a reduction in pressure. At lower altitudes the turning causes compression
shock waves (represented as solid lines), which causes local pressures to rise.
The compression waves are reflected off the boundary between the hot gas jet
and the ambient airstream, creating further expansion waves. At high altitude
there are no compression waves emanating from the ramp and the expanding
diverging flow exerts a decreasing pressure on the ramp area and behaves very
similarly to the diverging nozzle flow in a bell-shaped conventional nozzle exit
section. The contact locations, where the compression waves touch the ramp
surface, have higher local heat transfer than other areas on the ramp surface; these

8.3. LOW-THRUST ROCKETTHRUST CHAMBERS OR THRUSTERS 301

locations move downstream as the vehicle gains altitude. The wave patterns and
the pressure distribution on the spike or ramp surface can be determined from
computerized fluid dynamics programs or a method of characteristics program.

The advantages claimed for an aerospike engine are these: (1) Compared to
rocket engines with conventional axisymmetric nozzles, the truncated aerospike
is very short (about one third the length), and this also reduces the vehicle length
and vehicle structure mass; (2) it has altitude compensation and thus operates at
optimum nozzle expansion and highest possible exhaust velocity at every altitude;
and (3) differential throttling of certain sets of individual thruster modules or cells
allows pitch and yaw control, and with two engines it also allows roll control of
the vehicle during powered flight, as explained in Chapter 18. There is no gimbal
joint, no movement of the nozzle, no actuators, and no actuator power supply or
strong structural locations for actuator side loads. The disadvantages include the
lack of proven flight experience, proven reliability and performance validation,
and a larger- than-usual surface area subject to high heat transfer. There are some
small performance losses as the supersonic gas flow goes from a round nozzle
throat of each cell into a rectangular nozzle exit of the cell, also in turning the
flow on the ramp, and in combining supersonic flow from adjacent cell nozzles.
The average total loss with liquid propellants is believed to be about 1% of
the specific impulse. For combustion gases, which contain small particulates
(e.g., gases from solid composite propellant explained in Chapter 13), the losses
will be higher — up to 7% — depending on the percentage of particulates (which
are tiny solid particles and/or small liquid droplets). Due to the inertia of the
particulates, they can only be turned through a small angle in a high-speed gas
flow and they will hit the nozzle walls. The mass flow distribution will be uneven
with the heavy particulates mostly in the gas layers next to the concave curved
nozzle walls. Ref. 8-8 gives an example, which did not recognize the loss from
particulates.

8.3. LOW-THRUST ROCKET THRUST CHAMBERS OR THRUSTERS

Spacecraft, certain tactical missiles, missile defense vehicles, and upper stages of
ballistic missiles often use special, multiple thrusters in their small, liquid pro-
pellant rocket engines. They generally have thrust levels between about 0.5 and
10,000 N or 0. 1 to 2200 Ibf, depending on vehicle size and mission. As mentioned
in Section 4.5, they are used for trajectory corrections, attitude control, docking,
terminal velocity control in spacecraft or ballistic missiles, divert or side move-
ment, propellant settling, and other functions. Most operate with multiple restarts
for relatively short durations during a major part of their duty cycle. As mentioned
in Section 6.7, they can be classified into hot gas thrusters (high-performance
bipropellant with combustion temperatures above 2600 K and vacuum of 230
to 325 sec), warm gas thrusters such as monopropellant hydrazine (temperatures
between 500 and 1600 K and /, of 180 to 245 sec), and cold gas thrusters such
as high-pressure stored nitrogen (200 to 320 K) with low specific impules (40 to
120 sec).

302 THRUST CHAMBERS

A typical small thruster for bipropellant is shown in Fig. 8-13 and for
hydrazine monopropellant in Fig. 8-14. For attitude control angular motions
these thrust chambers are usually arranged in pairs as explained in Section 4.5.
The same control signal activates the valves for both units of such a pair. All
these small space rocket systems use a pressurized feed system, some with pos-
itive expulsion provisions, as explained in Section 6.2. The vehicle mission and
the automatic control system of the vehicle often require irregular and frequent
pulses to be applied by pairs of attitude control thrust chambers, which often
operate for very short periods (as low as 0.01 to 0.02 sec). This type of frequent
and short-duration thrust application is also known as pulsed thrust operation.
For translation maneuvers a single thruster can be fired (often in a pulsing mode)
and the thrust axis usually goes through the center of gravity of the vehicle. The
resulting acceleration will depend on the thrust and the location of the thruster
on the vehicle; it can be axial or at an angle to the flight velocity vector.

There is a performance degradation with decreasing pulse duration for all three
types of thrusters, because propellants are used inefficiently during the buildup
of thrust and the decay of thrust, when they operate below full chamber pressure
and the nozzle expansion characteristics are not optimum. The specific impulse

Niobium chamber with disilicide coating (0.003 in. thick)

2 Propellant valves

FIGURE 8-13. This radiation-cooled, insulated vernier thruster is one of several used
on the reaction control system of the Space Shuttle vehicle for orbit stabilization and
orientation, rendezvous or docking maneuvers, station keeping, deorbit, or entry. The
nozzle is cut off at an angle to fit the contour of the vehicle. Performance data are given
in Table 6-5. Operation can be in a pulse mode (firing durations between 0.08 and
0.32 sec with minimum offtime of 0.08 sec) or a steady-state mode (0.32 to 125 sec).
Demonstrated life is 23 hr of operation and more than 300,000 starts. (Courtesy of Aerojet
Propulsion Company and NASA.)

8.3. LOW-THRUST ROCKETTHRUST CHAMBERS OR THRUSTERS 303

Shower head

FIGURE 8-14. Typical hydrazine monopropellant small thrust chamber with catalyst
bed, showing different methods of injection.

really suffers when the pulse duration becomes very short. In Section 3.5 the
actual specific impulse of a rocket operating at a steady state was given at about
92% of theoretical specific impulse. With very short pulses (0.01 sec) this can
be lower than 50%, and with pulses of 0.10 sec it can be around 75 to 88%.
Also, the reproducibility of the total impulse delivered in a short pulse is not
as high after prolonged use. A preheated monopropellant catalyst bed will allow
performance improvement in the pressure rise and in short pulses.

One way to minimize the impulse variations in short pulses and to maximize
the effective actual specific impulse is to minimize the liquid propellant passage
volume between the control value and the combustion chamber. The propellant
flow control valves for pulsing attitude control thrust chambers are therefore
often designed as an integral part of the thrust chamber-injector assembly, as
shown in Fig. 8-13. Special electrically actuated leakproof, fast-acting valves
with response times ranging from 2 to 25 msec for both the opening and closing

304 THRUST CHAMBERS

operation are used. Valves must operate reliably with predictable characteristics
for perhaps 40,000 to 80,000 starts. This in turn often requires endurance proof
tests of 400,000 to 800,000 cycles.

Throttling or reducing the thrust is equivalent to reducing the total impulse
over the operating duration. This can be achieved by varying (1) the time between
pulses (less total cumulative impulse per unit time) or (2) by limiting the total
number of pulses.

Liquid storable bipropellants such as N204-monomethylhydrazine are used
when high performance is mandatory. Some have used ablative materials for
thrust chamber construction, as in the Gemini command module. The Space
Shuttle small thrusters use radiation cooling with refractory metals, as shown
in Fig. 8-13. A radiation-cooled thruster is shown later in Fig. 8-16. Carbon
materials made of woven strong carbon fibers in a matrix of carbon have also
been used in radiation-cooled bipropellant thrusters.

Hydrazine monopropellant thrusters are used when system simplicity is impor-
tant and moderate performance is acceptable. They have a nontoxic, clear, clean
exhaust plume. Virtually all hydrazine monopropellant attitude control rockets
use finely dispersed iridium or cobalt deposited on porous ceramic (aluminum
oxide) substrate pellets 1.5 to 3 mm in diameter as a catalyst. Figure 8-14 shows
a typical design of the catalyst pellet bed in an attitude control thruster designed
for pulse and steady-state operation meeting a specific duty cycle. Each injection
passage is covered with a cylindrical screen section which extends into a part of
the catalyst bed and distributes the hydrazine propellant. Figure 8-14 also shows
other successful types of hydrazine injector. Several arrangements of catalyst
beds are employed; some have spring-loading to keep the pellets firmly packed.
Hydrazine monopropellant thrust units range in size from 0.2 to 2500 N of thrust;
the scaling procedure is empirical and each size and design requires testing and
retesting. The amount of ammonia decomposition, shown in Fig. 7-3, can be
controlled by the design of the catalyst bed and its decomposition chamber.

Mechanical, thermal, and chemical problems arise in designing a catalyst bed
for igniting hydrazine, the more important of which are catalytic attrition and
catalyst poisoning. Catalytic attrition or physical loss of catalyst material stems
from motion and abrasion of the pellets, with loss of very fine particles. Crushing
of pellets can occur because of thermal expansion and momentary overpressure
spikes. As explained in Chapter 7, the catalyst activity can also decline because of
poisoning by trace quantities of contaminants present in commercial hydrazine,
such as aniline, MMH, UDMH, sulfur, zinc, sodium, or iron. Some of these
contaminants come with the hydrazine and some are added by the materials
used in the tanks, and propellant plumbing in the spacecraft. The high-purity
grade hydrazine has less than 0.003% aniline and less than 0.005% carbonaceous
material; it does not contaminate catalysts. Catalyst degradation, regardless of
cause, produces ignition delays, overpressures, and pressure spikes, decreases
the specific impulse, and decreases the impulse bit per pulse in attitude control
engines.

8.4. MATERIALS AND FABRICATION 305

Figure 17-4 shows a combination of chemical monopropellant and electrical
propulsion. Electrical postheating of the reaction gases from catalysis allows an
increase of the altitude specific impulse from 240 sec to about 290 or 300 sec.
A number of these combination auxiliary thrusters have successfully flown on a
variety of satellite applications and are particularly suitable for spacecraft where
electrical power is available and extensive short-duration pulsing is needed.

Cold gas thrusters use stored high-pressure inert gas as the propellant; they
and their performance were mentioned in Section 7.6 and their propellants and
specific impulses are listed in Table 7-3. They are simple, low cost, used with
pressurized feed systems, used for pulsing operations, and for low thrust and low
total impulse. They can use aluminum or plastics for thrusters, valves, and piping.
The early versions of the Pegasus air-launched launch vehicle used them for roll
control. The advantages of cold gas systems are: (a) they are very reliable and
have been proven in space flights lasting more than 10 years; (b) the system is
simple and relatively inexpensive; (c) the ingredients are nontoxic; (d) no deposit
or contamination on sensitive spacecraft surfaces, such as mirrors; (e) they are
safe; and (f) capable of random pulsing. The disadvantages are: (a) engines are
relatively heavy with poor propellant mass fractions (0.02 to 0.19); (b) the specific
impulses and vehicle velocity increments are low, when compared to mono- or
bipropellant systems; and (c) relatively large volumes.

8.4. MATERIALS AND FABRICATION

The choice of the material for the inner chamber wall in the chamber and the
nozzle throat region, which are the critical locations, is influenced by the hot
gases resulting from the burning of propellants, the maximum wall temperature,
the heat transfer, and the duty cycle. A large variety of material has been used.
Table 8-3 lists typical selected materials for several applications, thrust sizes,
and propellants. For high-performance, high heat transfer, regeneratively cooled
thrust chambers a material with high thermal conductivity and a thin-wall design
will reduce the thermal stresses. Copper is an excellent conductor and it will not
really oxidize in fuel-rich noncorrosive gas mixtures. It is often used with oxygen
and hydrogen below a mixture ratio of 6.0. The inner walls are therefore usually
made of a copper alloy (with small additions of zirconium, silver, or silicon),
which has a conductivity not quite as good as pure (oxygen-free) copper but has
improved high-temperature strength.

Figure 8-15 shows a partial section through five configurations of cooling
jackets. All involve curved or doubly curved components and precision fits for
good fastening or joining. The top configuration has an intermediate thin corru-
gated sheet of metal and has been used extensively in Russia and its predecessor
the Soviet Union. It is soldered together under external pressure in a furnace. It
is used in those locations on a thrust chamber, where the heat transfer is mild
or of intermediate intensity. The milled slot design can usually accept the high-
est heat transfer intensities. One fabrication technique is to machine (usually on

306 THRUST CHAMBERS

a milling machine) nearly rectangular grooves or cooling channels of varying
width and depth into the inner surface of a relatively thick contoured inner wall
using a high conductivity material such as a copper alloy. The grooves are then
filled with wax and by an electrolytic plating technique build an outer wall of a
suitable metal (such as nickel) so as to enclose the coolant channels. The wax
is then melted out. The third design shows a tubular construction; it has been
used extensively in the United States for larger thrust chambers with intermediate
to high heat transfer. Figure 8-9 shows that individual tubes are shaped to the
contour of the thrust chamber including its nozzle; the round tube shape is altered
into nearly rectangular cross sections. The formed and brazed tubes are brazed
or soldered together in a furnace and the outer structural shell or outer bands are
also brazed to the assembly. The corrugated outer wall concept is perhaps the
simplest and often the lightest cooling jacket configuration. It has been used in
locations where heat transfer is modest. The bottom shell represents a welded
construction, made of stainless steel. It has a higher allowable maximum wall
temperature than a brazed construction. It has been used in nozzles designed and
built in Europe.

Outer wall

Intermediate corrugated sheet
Inner wall

(a) intermediate corrugated brazed
sheet forms the cooling channels.

Outer shell bonded to tops
of channel walls

Milled cooling channel
Inner wall

(b) Milled channel design.

Outer struaural bands or shell
Tubes formed
Cooling passage

Corrugated outer wall

Cooling channel

inner wall

(d) Corrugated outer wail.

(e) Welded formed "Z"-shaped channels.

FIGURE 8-15. Simplified sketches of sections through a portion of the cooling jacket
of several different cooling schemes in regeneratively cooled thrust chambers. (Used with
permission from Ref. 8-1.)

8.4. MATERIALS AND FABRICATION 307

TABLE 8-3. Typical Materials Used in Several Common Liquid Propellant Thrust
Chambers

Cooling

Application

Propellant

Components

Method

Typical Materials

Bipropellant TC,

Oxygen —

C, N, E

Copper alloy

cooled, high
pressure (booster or
upper stage)

hydrogen

Transpiration cooled

porous stainless
steel face. Structure
is stainless steel

Alternate E

Carbon fiber in a

carbon matrix, or
niobium

Alternate E

Steel shell with

ablative inner liner

Bipropellant TC

Oxygen-

C, N, E, I

Stainless steel with

cooled, high

hycrocarbon

tubes or milled

pressure (booster or

or storable

slots

upper stage)

propellant"

Alternate E

Carbon fiber in a

carbon matrix, or
niobium

Alternate E

Steel shell with

ablative inner liner

Experimental TC

All types

C, N, E

Low carbon steel

(very limited
duration — only a
few seconds)

Small bipropellant TC

All types

C, N, E

Carbon fiber in

carbon matrix,
rhenium, niobium

Steel shell with

ablative inner
linear

Stainless steels.

titanium

Small monopropellant

Hydrazine

C, N, E,

Inconel, alloy steels

Stainless steel

Cold gas TC

Compressed

C, N, E, I

Aluminum, steel, or

air, nitrogen

plastic

“HNOs or N2O4 oxidizer with N2H4, MMH, or UDMH as fuels (see Chapter 7 ). TC = thrust
chamber, C = chamber wall, N = nozzle converging section walls, diverging section walls, and
throat region walls, E = walls at exit region of diverging section of nozzle, I = injector face, F = fuel
cooled (regenerative), R = radiation cooled, U = uncooled, T = transient heat transfer or heat sink
method (ablative material).

308 THRUST CHAMBERS

The depth and width of the cooling channels vary with the chamber pro-
hle location and the local thrust chamber diameter. At the nozzle throat region
the heat transfer is the highest; therefore it is the location of the highest cool-
ing velocity; therefore the total cooling passage area at the throat region is the
smallest. Often two or sometimes three different cooling jacket constructions are
designed into the same thrust chamber. Typically, the milled groove configuration
is used for the nozzle throat region and other configurations are then used for the
chamber and the nozzle exit regions.

The failure modes often are bulging on the hot wall on the gas side and the
opening up of cracks. During hot firing, fhe strain at the hot surface can exceed
the local yield point, thus giving it a local permanent compressive deformation.
With the cooldown after operation and with each successive firing, some addi-
tional yielding and further plastic deformation will occur until cracks form. With
successive firings the cracks can become deep enough for a leak of cooling pro-
pellant into the chamber or nozzle and the thrust chamber will then usually fail.
The useful life of a metal thrust chamber is the maximum number of firings
(and sometimes also the cumulative bring duration) without such a failure. The
prediction of wall failures is not simple and Refs. 8-5 and 8-6 explain this in
more detail. Useful life can also be limited by the storage life of soft compo-
nents (0-rings, gaskets, valve stem lubricant) and, for small thrusters with many
pulses, also the fatigue of valve seats. Therefore, there is a maximum limit on
the number of brings that such a thrust chamber can withstand safely, and this
limits its useful life.

For radiation cooling, several different carbon materials have worked in a
reducing, fuel-rich rocket combustion atmosphere. At leaner gas mixtures the
carbon can oxidize at the elevated temperatures when they glow red or white.
They can be used at wall temperatures up to perhaps 3300 K or 6000°R. Carbon
materials and ablative materials are used extensively in solid propellant rocket
motors and are discussed further in Chapter 15.

For some small radiation-cooled bipropellant thrusters with storable propel-
lants, such as those of the reaction control thrusters on the Space Shuttle orbiter,
the hot walls are made of niobium coated with disilicide (up to 1120 K or
2050°R). To prevent damage, a fuel-rich mixture or blm cooling is often used.
For small thrusters rhenium walls protected by iridium coatings (oxidation resis-
tant) have come into use more recently and have been used up to about 2300 K or
4100°R (see Refs. 8-1 and 8-9). Other high-temperature materials, such as tung-
sten, molybdenum, alumina, or tantalum, have been tried but have had problems
in manufacture, cracking, hydrogen embrittlement, or excessive oxidation.

A small radiation-cooled bipropellant thruster is shown in Fig. 8-16. It uses
three different nozzle and chamber materials. This thruster’s injector has extra
fuel injection holes (not shown in Fig. 8-16) to provide blm cooling to keep wall
temperatures below their failure limits. High temperature copper-nickel alloys or
stainless steels are used for the radiation cooled nozzle-and chamber walls of
thrusters operating with hydrazine monopropellant. See Fig. 8-14.

8.4. MATERIALS AND FABRICATION 309

FIGURE 8-16. Radiation-cooled reaction control thruster R-4D-15 uses nitrogen tetrox-
ide and monomethylhydrazine propellants. The large nozzle area ratio allows good vacuum
performance. It has three different nozzle materials, each with a lower allowable temper-
ature (Re 4000°F; Nh 3500°F; Ti 1300°F). Courtesy of Aerojet Propulsion Company.

Until recently it has not been possible to make large pieces of carbon-carbon
material. This was one of the reasons why large nozzle sections and integral
nozzle exit cone pieces in solid motors were made from carbon phenolic cloth
lay-ups. Progress in manufacturing equipment and technology has now made it
possible to build and fly larger pieces made of carbon fiber in a carbon matrix. A
three-piece extendible carbon nozzle exit cone of 2.3 m (84 in.) diameter and 2.3
to 3 mm thickness has flown on an upper-stage engine. This thrust chamber with
its movable nozzle extension is shown in Fig. 8-17, its parameters are listed in
Table 8-1, its testing is reported in Ref. 8-4, and it has flown successfully many
times.

The material properties have to be evaluated under all likely operating condi-
tions, loads, start conditions, pressure variations, and the like before a material
can be selected for a specific thrust chamber application. This evaluation includes
physical properties, such as tensile and compressive strengths, yield strength, frac-
ture toughness, modulus of elasticity (for determining deflections under load),
thermal conductivity (a high value is best for steady-state heat transfer), coef-
ficient of thermal expansion (some large thrust chambers grow by 3 to 10 mm
when they become hot, and that can cause problems with their piping connec-
tions or structural supports), specific heat (capacity to absorb thermal energy),
reflectivity (for radiation), or density (ablatives require more volume than steel).

310 THRUST CHAMBERS

Nozzle extension device (motors)

Gimbal mount

Controller

Regeneratively
cooled chamber
and nozzle

Propellant

cooldown

lines

Movable
nozzle extension

(a) Half section of nozzle extension in stowed position

FIGURE 8-17. The RL-lOB-2 rocket engine has an extendible nozzle cone or skirt,
which is placed around the engine during the ascent of the Delta 111 and IV launch
vehicles. This extension is lowered into position by electromechanical devices after the
launch vehicle has been separated from the upper stage at high altitude and before firing.
(Courtesy of Pratt & Whitney Rocketdyne, of United Technologies, Inc.)

8.5. HEAT TRANSFER ANALYSIS 311

These properties change with temperature (they are different when they are hot)
and sometimes they change with little changes in materials composition. The
temperature where a material loses perhaps 60 to 75% of its room temperature
strength is often selected as the maximum allowable wall temperature, well below
its melting point. Since a listing of all the key properties of a single material
requires many pages, it is not possible to list them here, but they are usually
available from manufacturers and other sources. Other important properties are
erosion resistance, little or no chemical reactions with the propellants or the hot
gases, reproducible decomposition or vaporization of ablative materials, ease and
low cost of fabrication (welding, cutting, forming, etc.), the consistency of the
composition (impurities) of different batches of each material (metals, organics,
seals, insulators, lubricants, cleaning fluids), and ready availability and low cost
of each material.

8.5. HEAT TRANSFER ANALYSIS

In actual rocket development not only is the heat transfer analyzed but the rocket
units are almost always tested to assure that heat is transferred satisfactorily under
all operating and emergency conditions. Heat transfer calculations are useful to
guide the design, testing, and failure investigations. Those rocket combustion
devices that are regeneratively cooled or radiation cooled can reach thermal equi-
librium and the steady-state heat transfer relationships will apply. Transient heat
transfer conditions apply not only during thrust buildup (starting) and shutdown
of all rocket propulsion systems, but also with cooling techniques that never
reach equilibriurm, such as with ablative materials.

Sophisticated finite element analysis (FEA) programs of heat transfer have
been available for at least 20 years, and several different FEA computer programs
have been used for the analysis of thrust chamber steady-state and transient
heat transfer, with different chamber geometries or different materials and with
temperature-variant properties. A detailed description of this powerful analysis
is beyond the scope of this book, but can be found in Refs. 8-10, 8-11, and
8-12. Major rocket propulsion organizations have developed their own versions
of suitable computer programs for solving their heat transfer problems. This
section gives the basic relationships that are the foundation for FEA programs.
They are intended to give some understanding of the phenomena and underlying
principles.

General Steady-State Heat T ransfer Relations

For heat transfer conduction the following general relation applies:

Q dT AT

(8-13)

312 THRUST CHAMBERS

where Q is the heat transferred across a surface area A, dT/dL the temperature
gradient with respect to thickness at the surface A, and k the thermal con-
ductivity expressed as the amount of heat transferred per unit time through a
unit area of surface for 1° temperature difference over a unit wall thickness. The
negative sign indicates that temperature decreases as thickness increases.

The steady-state heat transfer through the chamber wall of a liqud-cooled
rocket chamber can be treated as a series-type, steady-state heat transfer problem
with a large temperature gradient across the gaseous film on the inside of the
chamber wall, a temperature drop across the wall, and, in cases of cooled cham-
bers, a third temperature drop across the film of the moving cooling fluid. It is a
combination of convection at the boundaries of the flowing fluids and conduction
through the chamber walls. It is shown schematically in Fig. 8-18.

The general steady-state heat transfer equations for regeneratively cooled thrust
chambers can be expressed as follows:

h{Tg-Ti) = Q/A

(8-14)

Tg-Ti

l/hg + t„ 1 K -F l/hi

(8-15)

(8-16)

{K/t„)(T„g - T„i)

(8-17)

hi{T„i — Ti)

(8-18)

where q is heat transferred per unit area per unit time, Tg the absolute chamber
gas temperature, T/ the absolute coolant liquid temperature, T^,i the absolute wall
temperature on the liquid side of the wall, T^g the absolute wall temperature on

FIGURE 8-18. Temperature gradients in cooled rocket thrust chamber. The listed tem-
peratures are typical.

8.5. HEAT TRANSFER ANALYSIS 313

the gas side of the wall, h the overall film coefficient, hg the gas film coefficient,
hi the coolant liquid film coefficient, the thickness of the chamber wall, and
K the conductivity of the wall material. The strength and thermal properties of
materials are functions of temperature. Any consistent set of units can be used
in these equations. These simple relations assume that the heat flow is radial.
The simple quasi-one-dimensional theory also often assumes that the thermal
conductivity and the film coefficients are at average values and not functions
of temperature or pressure. A two- or three-dimensional finite element model
would also need to be used to analyze the heat transfer in the axial directions,
which usually occurs in the nozzle throat wall regions; some of the heat from
the hot nozzle throat is transferred to wall regions upstream and downstream of
the throat.

Because the film coefficients, the gas and liquid coolant temperatures, the wall
thickness, and the surface areas usually vary with the axial distance L within
a combustion chamber (assuming axial heat transfer symmetry), the total heat
transfer per unit time Q can be found by integrating the local heat transfer over
the entire internal surface area of the chamber and the nozzle:

Because both q and D are complicated functions of the thrust chamber length
L, the equation usually has to be solved by dividing the rocket chamber into finite
lengths. Assuming that q is given by Eqs. 8-13 to 8-19 and remains constant
over the length of each element gives an approximate solution.

The important quantities for controlling the heat transfer across a rocket cham-
ber wall are the fluid film boundaries established by the combustion products on
one side of the wall and the coolant flow on the other. The gas film coefficient
largely determines the numerical value of the heat transfer rate, and the liquid
film largely determines the value of the wall temperatures. The determination of
the film coefficients in Eqs. 8-16 and 8-18 is difficult because of the complex
geometries, the nonuniform velocity profile, the surface roughness, the boundary
layer behavior, and the combustion oscillations.

Conventional heat transfer theory is usually given in terms of several dimen-
sionless parameters (Ref. 8-10, 8-11, and 8-12):

where hg is the film coefficient, D the diameter of the chamber of the nozzle,
V the calculated average local gas velocity, ic the conductivity of the gas, /u, the
absolute gas viscosity, Cp the specific heat of the gas at constant pressure, and p
the gas density.

In Eq. 8-20 the quantity hgD /k is known as the Nusselt number, the quantity
Dvp/ pL as the Reynolds number, and the quantity Cppt/K as the Prandtl number

(8-19)

( 8 - 20 )

314 THRUST CHAMBERS

Pr. The gas film coefficient hg can be determined from Eq. 8-21:

( 8 - 21 )

where pn is the local mass velocity, and the constant 0.026 is dimensionless. The
boundary layer temperature gradient has effects on the various gas properties in
rocket combustion, Conventional theoretical approaches, such as for Eq. 8-20
or 8-21, are usually for heated circular tubes, relatively long tubes, steady-state
flow, and they reach an equilibrium velocity profile. Heat flows into the tube from
all sides (360°). However, in thrust chambers the heat flow to the coolant passage
is only on one side of the cooling passage and the combustion phenomena are
propellant specific. The chamber length is relatively short, and an equilibrium
flow profile does not exist. The actual flow in the combustion chamber is highly
turbulent. There are liquid droplets, which evaporate, and there is no equilibrium.
For these reasons Eqs. 8-20 and 8-21 an approximations only.

Equations where the coefficients have been validated by actual experimen-
tal data are more reliable and they are used in design. Bartz (Ref. 8-13) has
surveyed the agreement between theory and experiment and developed semiem-
pirical correction factors:

The subscript 0 refers to properties evaluated at the stagnation or combustion
temperature; the subscript am refers to properties at the arithmetic mean temper-
ature of the local free-stream static temperature and the wall temperatures; and
p' is the free-stream value of the local gas density. Again, the empirical constant
0.026 is dimensionless when compatible dimensions are used for the other terms.
The gas velocity v is the local free-stream velocity corresponding to the density
p'. Since density raised to the 0.8 power is roughly proportional to the pressure
and the gas film coefficient is roughly proportional to the heat flux, it follows that
the heat transfer rate increases approximately linearly with the chamber pressure.
These semiempirical heat transfer equations have been modified and validated
for common propellants, limited chamber pressure ranges, and specific injectors
(see Ref. 8-13) and are often proprietary to specific design organizations.

The temperature drop across the inner wall and the maximum temperature are
reduced if the wall is thin and is made of material of high thermal conductiv-
ity. The wall thickness is determined from strength considerations and thermal
stresses, and some designs have as little as 0.025 in. thickness. The effect of
changing the film coefficients is shown in Example 8-1.

Surface roughness can have a large effect on the film coefficients and thus on
the heat flux. Measurements have shown that the heat flow can be increased by
a factor of up to 2 by surface roughness and to higher factors when designing
turbulence-creating obstructions in the cooling channels. Major surface roughness

( 8 - 22 )

8.5. HEAT TRANSFER ANALYSIS 315

on the gas side will cause the gas locally to come close to stagnation temper-
ature. However, surface roughness on the liquid coolant side of the wall will
enhance turbulence and the absorption of heat by the coolant and reduce wall
temperatures.

Example 8-1. The effects of varying the film coefficients on the heat transfer and the
wall temperatures are to be explored. The following data are given:

Wall thickness 0.445 mm

Wall material Low-carbon steel

Average conductivity 43.24 W/m^- K/m

Average gas temperature 3033 K or 2760°C

Average liquid bulk temperature 311.1 K or 37.8°C

Gas-film coefficient 147 W/m^-°C

Liquid-film coefficient 205,900 W/m^-°C

Vary hg (at constant hi), then vary hi (at constant hg), and then determine the changes in
heat transfer rate and wall temperatures on the liquid and the gas side of the wall.

SOLUTION. Use Eqs. 8-13 to 8-18 and solve for q, T„g, and The answers shown
in Table 8-4 indicate that variations in the gas-film coefficient have a profound influence
on the heat transfer rate but relatively little effect on the wall temperature. The exact
opposite is true for variations in the liquid-film coefficient; here, changes in hi produce
little change in q but a fairly substantial change in the wall temperature.

TABLE 8-4. Change in Film Coefficient for Example 8-1

Change in Film
Coefficient (%)

Change in
Heat Transfer

(%)

Wall Temperature (K)

Gas Film

Liquid Film

Gas Side, Ty„g

Liquid Side, T^i

100

324.4

321.1

100

337.2

330.5

200

100

198

362.8

349.4

400

100

389

415.6

386.1

100

356.1

349.4

100

393.3

386.7

100

12.5

460.0

397.8

100

6.25

596.7

590.5

Figure 8-19 shows heat flow directions, temperature distribution, and the
locations for the maximum wall temperatures for a milled cooling jacket design.
This design is represented by the third sketch of Fig. 8-15. The inner wall should
be thin, so that the temperature difference across this wall is low and therefore
the thermal stresses are also low.

316 THRUST CHAMBERS

Estimated

temperature

distribution

Hottest areas

Isotherms (solid lines)

Dashed lines indicate
direction of heat flux

FIGURE 8-19. Results of a two-dimensional analysis of the heat transfer in two cooling
channels of a milled slot cooling jacket. The outer wall and the upper parts of the channels
are not shown.

T ransient Heat T ransfer Analysis

An uncooled (high melting point) metal thrust chamber is the simplest type to
analyze because there is no chemical change. Thermal equilibrium is not reached.
The uncooled walls act essentially as a heat sponge and absorb heat from the hot
gases. With the aid of experimental data to determine some typical coefficients,
it is possible in some cases to predict the transient heating of uncooled walls.

Heat is transferred from the hot gases to the wall, and during operation a
changing temperature gradient exists across the wall. The heat transferred from
the hot wall to the surrounding atmosphere, and by conduction of metal parts to
the structure, is negligibly small during this transient heating. Each local point
within the wall has its temperature raised as the burning process is extended in
time. After the completion of the rocket’s operation, the wall temperatures tend
to equalize. A typical temperature -time -location history is given in Fig. 8-20.
Here the horizontal line at T = 21°C denotes the initial equilibrium condition
of the wall before the rocket operates; the various curves show the temperature
profile across the wall at successive time intervals after initiation of combustion.
The line at T = 357°C shows an equilibrium temperature of the wall a finite time
after cutoff.

The heat transferred across the hot surface of the wall (and distributed within
the wall by conduction) must be less than the heat-absorbing capacity of the wall
material below the critical temperature. If heat transfer to the outside atmosphere
and axially within the metal wall is neglected, this can be expressed in a simplified
form:

g At = -KA{dT/dL)At = me AT (8-23)

where Q is the heat per second transferred across an inner wall surface area A.
Equation 8-14 shows that Q/A depends on the hot gas temperature, the wall
temperature, and the gas film coefficient. The heat conductivity k depends on the
material and its temperature; AT denotes the average wall temperature increment;
dT/dL the temperature gradient of the heat flow near the hot wall surface in
degrees per unit thickness; m the mass of a unit area of wall; c the average
specific heat of the wall material; and At is the time increment. The chamber and
nozzle walls can be divided into cylindrical or conical segments, and each wall

8.5. HEAT TRANSFER ANALYSIS 317

Wall position number

Wall thickness, 12.7 mm H

Heated Insulated

surface surface

FIGURE 8-20. Typical temperature distributions through a wall of an uncooled relatively
thick metal thrust chamber as a function of heating time.

segment in turn is divided into an arbitrary number of axisymmetric concentric
layers, each of a finite thickness. At any given time the heat conducted from any
one layer of the wall exceeds the heat conducted into the next outer layer by
the amount of heat absorbed in raising the temperature of the particular layer.
This iterative approach lends itself readily to two- or three-dimensional computer
analysis, resulting in data similar to Fig. 8-20. It is usually sufficient to determine
the heat transfer at the critical locations, such as in the nozzle throat region.

318 THRUST CHAMBERS

A more complex three-dimensional analysis can also be undertaken; here the
wall geometry is often more complex than merely cylindrical, heat is conducted
also in directions other than normal to the axis, temperature variable properties
are used, boundary layer characteristics vary with time and location, and there
may be more than one material layer in the wall.

A number of mathematical simulations of transient heat transfer in ablative
materials have been derived, many with limited success. This approach should
include simulation for the pyrolysis, chemical decomposition, char depth, and
out-gassing effects on film coefficient, and it requires good material property
data. Most simulations require some experimental data.

Steady-State Transfer to Liquids in Cooling Jacket

The term regenerative cooling is used for rockets where one of the propellants is
circulated through cooling passages around the thrust chamber prior to the injec-
tion and burning of this propellant in the chamber. It is really forced convection
heat transfer. The term regenerative is perhaps not altogether appropriate here,
and it bears little relation to the meaning given to it in steam turbine practice. It
is intended to convey the fact that the heat absorbed by the coolant propellant is
not wasted but augments its initial temperature and raises its energy level before
it passes through the injector. This increase in the internal energy of the liquid
propellant can be calculated as a correction to the enthalpy of the propellant
(see Chapter 5). However, the overall effect on rocket performance is usually
very slight. With some propellants the specific impulse can be 1% larger if the
propellants are preheated through a temperature differential of 100 to 200° C.
In hydrogen-cooled thrust chambers and in small combustion chambers, where
the wall-surface-to-chamber volume ratio is relatively large, the temperature rise
in the regenerative coolant will be high, and the resulting increase in specific
impulse is sometimes more than 1%.

The behavior of the liquid film is critical for controlling the wall temperatures
in forced convection cooling of rocket devices at high heat fluxes (see Table 8-4
and Refs. 8-10, 8-14, and 8-15). At least four different types of film appear
to exit, as can be interpreted from Fig. 8-21. Here the heat transfer rate per
unit of wall surface q is shown as a function of the difference between the wall
temperature on the liquid side T„i and the bulk temperature of the liquid T;.

1. The normal forced convection region at low heat flux appears to have
a liquid boundary layer of predictable characteristics. It is indicated by
region A-B in Fig. 8-21. Here the wall temperature is usually below the
boiling point of the liquid at the cooling jacket pressure. In steady-state
heat transfer analysis the liquid-film coefficient can be approximated by
the usual equation (see Refs. 8-10, 8-11, and 8-12):

(8-24)

8.5. HEAT TRANSFER ANALYSIS

319

FIGURE 8-21. Regimes in transferring heat from a hot wall to a flowing liquid.

where m is the liquid fluid mass flow rate, c its average specific heat,
A the cross-sectional cooling jacket flow area, D the equivalent diameter
of the coolant passage cross section,* v the fluid velocity, p the coolant
density, /r its absolute viscosity, and /c its conductivity. Many liquid-cooled
rocket devices operate in this regime of heat transfer. Values of the physical
properties of several propellants are given in Tables 8-5 and 7-1.

2. When the wall temperature T„i exceeds the boiling point of the liquid by
perhaps 10 to 50 K, small vapor bubbles form at the hot wall surface.
These small, nuclei-like bubbles cause local turbulence, break away from
the wall, and collapse in the cooler liquid. This phenomenon is known
as nucleate boiling. The turbulence induced by the bubbles changes the
character of the liquid film and, augmented by the vaporization of some
of the propellant, the heat transfer rate is increased without a proportional
increase in the temperature drop across the film, as can be seen by the steep
slope B-C of the curve in Fig. 8-21. If the pressure of the fluid is raised,
then the boiling point is also raised and the nucleate boiling region shifts to
the right, to B'-C'. This boiling permits a substantial increase in the heat
transfer beyond that predicted by Eq. 8-24. This phenomenon often occurs
locally in the nozzle throat area, where the heat flux is high. The maximum
feasible heat transfer rate (point C) is indicated as q^nx in Table 8-5 and

*The grooves, tubes, or coolant passages in liquid propellant rocket chambers are often of complex
cross section. The equivalent diameter, needed for fluid-film heat transfer calculations, is usually
defined as four times the hydraulic radius of the coolant passage; the hydraulic radius is the cross-
sectional flow area divided by the wetted perimeter.

320 THRUST CHAMBERS

appears to be a function of the cooling-fluid properties, the presence of
dissolved gases, the pressure, and the flow velocity. In Table 8-5 it can
be seen that hydrazine is a good heat absorber, but kerosene is poor (low
^max and low critical pressure).

3. As the heat transfer is increased further, the rate of bubble formation and
the bubble size become so great that the bubbles are unable to escape from
the wall rapidly enough. This reaction (shown as C-D in Fig. 8-21) is
characterized by an unstable gas film and is difficult to obtain reproducibly
in tests. When a film consisting largely or completely of gas forms along
the hot wall surface, then this film acts as an insulation layer, causing a
decrease in heat flux and, usually, a rapid increase in wall temperature,
often resulting in a burnout or melting of the wall material. The cooling
flow system has to be designed to avoid this unstable gas film regime

4. As the temperature difference across the film is further increased, the wall
temperatures reach values in which heat transfer by radiation becomes
important. Region D-E is not of interest to cooling jacket designers.

Cooling can also be accomplished by a fluid above its critical point with
coolants such as hydrogen. In this case there is no nucleate boiling and the heat
transfer increases with the temperature difference, as shown by the supercritical
(dashed) line in Fig. 8-21. Liquid hydrogen is an excellent coolant, has a high
specific heat, and leaves no residues.

Chemical changes in the liquid can seriously influence the heat transfer from
hot walls to liquids. Cracking of hydrocarbon fuel, with an attendant formation
of insoluble gas, tends to reduce the maximum heat flux and thus promote failure
more readily. Hydrocarbon fuel coolants (methane, jet fuel) can break down and
form solid, sticky carbon deposits inside the cooling channel, impeding the heat
transfer. Other factors influencing steady-state coolant heat transfer are gas radi-
ation to the wall, bends in the coolant passage, improper welds or manufacture,
and flow oscillations caused by turbulence or combustion unsteadiness. Some
propellants, such as hydrazine, can decompose spontaneously and explode in the
cooling passage if they become too hot.

To achieve a good heat-absorbing capacity of the coolant, the pressure and
the coolant flow velocity are selected so that boiling is permitted locally, but the
bulk of the coolant does not reach this boiling condition. The total heat rejected
by the hot gases to the surface of the hot walls, as given by Eq. 8-14 must be
less than that permitted by the temperature rise in the coolant, namely

qA = Q= mc(Ti - T 2 ) (8-25)

where m is the coolant mass flow rate, c the average specific heat of the liquid,
Ti the initial temperature of the coolant as it enters the cooling jacket, and T 2 its
final temperature; Q is the rate of heat absorption per unit time; q is this same
rate per unit heat transfer area A; T 2 should be below the boiling point prevailing
at the cooling jacket pressure.

CN CM ’-H (N ’-H

0\ CM r^
^

° -a

3 3

[C Dh

IT) IT)

oooo-Hco-Hodo

-H CM ^ CM CM

o o o

^ CM

»n

M3 O

<n cn

Q-

•c

Cl-

00 ^
c ir

bJ) ^

^ Oh
O C

CQ g

V5 Cl

V3 M

i> p

CM 00

in i>

M3 M3

^ CM

m CM

r^inoooom-H-^^cM^

ooln'^ooo^Olnlno^'^o^

m-^inin'^'sO'O^cMcnm

M3 o o^
cn o 00
m ^ ^

-H o^ ^ On -h o^ ^

oooinONOooooooooomo
-HNO^oo-HvomcO’-HNO^^

-H in

o ^

o o cn NO o o

o o ^ o -H cn

-a

I I

' ^

] X

I a

3 ^

321

322 THRUST CHAMBERS

Radiation

Radiation heat emission is the electromagnetic radiation emitted by a gas, liquid,
or solid body by the virtue of its temperature and at the expense of its inter-
nal energy. It covers the wavelength range from 10,000 to 0.0001 jam, which
includes the visible range of 0.39 to 0.78 p.m. Radiation heat transfer occurs
most efficiently in a vacuum because there is no absorption by the intervening
fluids.

The heat transmitted by the mechanism of radiation depends primarily on the
temperature of the radiating body and its surface condition. The second law of
thermodynamics can be used to prove that the radiant energy £ is a function of
the fourth power of the absolute temperature T :

E =fsaAT"^ (8-26)

The energy E radiated by a body is defined as a function of the emissivity e,
which is a dimensionless factor for surface condition and material properties,
the Stefan -Boltzmann constant cr(5.67 x 10“^ W/m^-K"^), the surface area A,
the absolute temperature T, and the geometric factor/, which depends on the
arrangement of adjacent parts and the shape. At low wall temperatures (below
800 K) radiation accounts for only a negligible portion of the total heat transfer
in a rocket device and can usually be neglected.

In rocket propulsion there are these radiation concerns:

1 . Emission of hot gases to the internal walls of a combustion chamber, and its
nozzle converging section, a solid propellant grain, or a hybrid propellant
grain.

2. Emission to the surroundings or to space from the external surfaces of
hot hardware (radiation-cooled chambers, nozzles, or electrodes in electric
propulsion).

3. Radiation from the hot plume downstream of the nozzle exit. This is
described in Chapter 20.

In rocket combustion devices gas temperatures are between 1900 and 3900 K
or about 3000 to 6600° F; their radiation contributes between 3 and 40% of the
heat transfer to the chamber walls, depending on the reaction gas composition,
chamber size, geometry, and temperature. It can be a significant portion of the
total heat transfer. In solid propellant motors the radiation heating of the grain
surfaces can be critical to the burning rate, as discussed in Chapter 12. The
absorption of radiation on the wall follows essentially the same laws as those of
emission. Metal surfaces and formed tubes reflect much of the radiant energy,
whereas ablative materials and solid propellants seem to absorb most of the
incident radiation. A highly reflective surface on the inside wall of a combustor
tends to reduce absorption and to minimize the temperature increase of the walls.

The hot reaction gases in rocket combustion chambers are potent radiation
sources. Gases with symmetrical molecules, such as hydrogen, oxygen, and

8.6. STARTING AND IGNITION 323

nitrogen, have been found not to show many strong emission bands in those
wavelength regions of importance in radiant heat transfer. Also, they do not
really absorb much radiation and do not contribute considerable energy to the
heat transfer. Heteropolar gases, such as water vapor, carbon monoxide, car-
bon dioxide, hydrogen chloride, hydrocarbons, ammonia, oxides of nitrogen, and
the alcohols, have strong emission bands of known wavelengths. The radiation of
energy of these molecules is associated with the quantum changes in their energy
levels of rotation and interatomic vibration. In general, the radiation intensity of
all gases increases with their volume, partial pressure, and the fourth power of
their absolute temperature. For small thrust chambers and low chamber pressures,
radiation contributes only a small amount of energy to the overall heat transfer.

If the hot reaction gases contain small solid particles or liquid droplets, then
the radiation heat transfer can increase dramatically by a factor of 2 to 10. The
particulates greatly increase the radiant energy as explained in Section 20.1. For
example, the reaction gas from a gelled liquid propellants contains small solid
particles and many solid propellants contain fine aluminum powder. When burned
to form aluminum oxide, the heat of combustion and the combustion temperature
are increased (raising heat transfer), and the specific impulse is raised somewhat
(giving improved performance). The oxide can be in the form of liquid droplets
(in the chamber) or solid particles (in the nozzle diverging section), depending on
the local gas temperature. Furthermore, the impact of these particulates with the
wall will cause an additional increase in heat transfer, particularly to the walls
in the nozzle throat and immediately upstream of the nozzle throat region. The
particles also cause erosion or abrasion of the walls.

8.6. STARTING AND IGNITION

The starting of a thrust chamber has to be controlled so that a timely and even
ignition of propellants is achieved and the flow and thrust are built up smoothly
and quickly to their rated value (see Refs. 6-1, 8-16, and 8-17). The initial
propellant flow is always less than full flow , and the starting mixture ratio is
usually different from the operating mixture ratio. A low initial flow prevents
strong water hammer, gives a low initial heat release, and for nonhypergolic
propellants it prevents an excessive accumulation of unignited liquid propellants
in the chamber.

The starting injection velocity is low, the initial vaporization, atomization, and
mixing of propellants in a cold combustion chamber is often incomplete, and
there are local regions of lean and rich mixtures. With cryogenic propellants the
initial chamber temperature can be below ambient. The optimum starting mixture
is therefore only an average of a range of mixture ratios, all of which should
be readily ignited. Mixture ratios near the stoichiometric mixture ratio have a
high heat release per unit of propellant mass and therefore permit bringing the
chamber and the gases up to equilibrium faster than would be possible with other
mixtures. The operating mixture ratio is usually fuel rich and is usually selected

324 THRUST CHAMBERS

for optimum specific impulse. One method of analytical modeling of the ignition
of cryogenic propellants is given in Ref. 8-16.

The time delay for starting a thmst chamber ideally consists of the following
time periods:

1. Time needed to fully open the propellant valves (typically 0.002 to more
than 1.00 sec, depending on valve type and its size and upstream pressure).

2. Time needed to fill the liquid passage volume between the valve seat and
the injector face (piping, internal injector feed holes, and cavities).

3. Time for forming discrete streams or sprays of liquid propellant (sometimes
gaseous propellant, if cryogenic liquid is preheated by heat from cooling
jacket) and for initial atomization into small droplets and for mixing these
droplets.

4. For today’s hypergolic propellant combinations at ambient initial tempera-
tures, the combustion starts within a few milliseconds after a fuel droplet
or fuel vapor comes into contact with an oxidizer droplet or oxidizer vapor.
This delay can be longer at low ambient temperatures and at off-mixture
ratios.

5. For nonhypergolic propellant combinations an igniter system has to provide
sufficient heat to bring a flow of mixed propellant to its ignition tempera-
ture, before combustion can safely start. The igniter is usually started before
any propellant is admitted to the chamber and usually before the propellant
valves are opened. The time for the igniter to operate before any propellant
comes in contact with the igniter flame can be one or more seconds in
larger thrust chambers. If the igniter fails to operate and if this is sensed,
then the engine controls can block the opening of the propellant valves.

6. Time needed for droplets to vaporize and ignite (laboratory tests show this
to be short, 0.02 to 0.05 sec, but this depends on the propellants and the
available heat).

7. Once ignition is achieved at a particular location in the chamber, it takes
time to spread the flame or to heat all the mixed propellant that is entering
into the chamber, vaporizing it, and raising it to ignition temperature.

8. Time needed to raise the chamber pressure and temperature to the point
where combustion will be self-sustaining. Then it is raised to its full pres-
sure.

There are overlaps in these delays and several of them can occur simultaneously.
The delays are longer with large injectors or large-diameter chambers. Small
thrusters can usually be started very quickly, in a few milliseconds, while larger
units require 1 sec or more, sometimes as much as 5 sec.

In starting a thmst chamber, one propellant always reaches the chamber a
short time ahead of the other; it is almost impossible to synchronize exactly
the fuel and oxidizer feed systems so that the propellants reach the chamber
simultaneously at all injection holes or spray elements. Frequently, a more reliable

8.6. STARTING AND IGNITION 325

ignition is assured when one of the propellants is intentionally made to reach the
chamber first. For example, for a fuel-rich starting mixture the fuel is admitted
first. Reference 8-17 describes the control of the propellant lead.

Other factors influencing the starting flows, the propellant lead or leg, and
some of the delays mentioned above are the liquid pressures supplied to the
injector (e.g., regulated pressure), the temperature of the propellant (some can be
close to their vapor point), and the amount of insoluble gas (air bubbles) mixed
with the initial quantity of propellants.

The propellant valves (and the flow passages betwen the valves and the injec-
tor face) are often so designed and controlled that they operate in a definite
sequence, thereby assuring an intentional lead of one of the propellants and a
controlled buildup of flow and mixture ratio. Often the valves are only partially
opened because it is easier to ignite a small flow and it avoids an accumulation
of hazardous unburned propellant mixture in the chamber. Once combustion is
established, the valves are fully opened and full flow may reach the thrust cham-
ber assembly. The initial reduced flow burning period is called the preliminary
stage. Section 11.4 describes the starting controls.

Full flow in the larger thrust chambers is not initiated with non- self-igniting
propellants until the controller received a signal of successful ignition. The verifi-
cation of ignition or initial burning is often built into engine controls using visual
detection (photocell), heat detection (pyrometer), a fusible wire link, or sensing
of a pressure rise. If the starting controls are not designed properly, unburnt
propellant may accumulate in the chamber; upon ignition it may then explode,
causing sometimes severe damage to the rocket engine. Starting controls and
engine calibrations are discussed in Sections 11.4 and 11.5.

Nonspontaneously ignitable propellants need to be activated by absorbing
energy prior to combustion initiation. This energy is supplied by the ignition
system. Once ignition has begun the flame is self-supporting. The igniter has to
be located near the injector in such a manner that a satisfactory starting mixture
at low initial flow is present at the time of igniter activation, yet it should not
hinder or obstruct the steady-state combustion process. At least five different
types of successful propellant ignition systems have been used.

Spark plug ignition has been used successfully on liquid oxygen -gasoline
and on oxygen -hydrogen thrust chambers, particularly for multiple starts during
flight. The spark plug is often built into the injector, as shown in Fig. 9-6.

Ignition by electrically heated wires has been accomplished but at times has
proven to be less reliable than spark ignition for liquid propellants.

Pyrotechnic ignition uses a solid propellant squib or grain of a few seconds’
burning duration. The solid propellant charge is electrically ignited and bums
with a hot flame within the combustion chamber. Almost all solid propellant
rockets and many liquid rocket chambers are ignited in this fashion. The igniter
container may be designed to fit directly onto the injector or the chamber (see
Fig. 8-1), or may be held in the chamber from outside through the nozzle. This
ignition method can only be used once; thereafter the charge has to be replaced.

326 THRUST CHAMBERS

Ignition has been achieved in a precombustion chamber also called premix
chamber; see Fig. 8-3; it is a small chamber built next to the main combustion
chamber and connected through an orifice; this is similar to the precombustion
chamber used in some internal combustion engines. A small amount of fuel and
oxidizer is injected into the precombustion chamber and ignited. The burning
mixture enters the main combustion chamber in a torchlike fashion and ignites
the larger main propellant flow which is injected directly into the main chamber.
This ignition procedure permits repeated starting of thrust chambers and has
proved successful with the liquid oxygen -gasoline and oxygen -hydrogen thrust
chambers.

Auxiliary fluid ignition is a method whereby some hypergolic liquid or gas, in
addition to the regular fuel and oxidizer, is injected into the combustion chamber
for a very short period during the starting operation. This fluid is hypergolic,
which means it produces spontaneous combustion with either the fuel or the oxi-
dizer. The combustion of nitric acid and some organic fuels can, for instance,
be initiated by the introduction of a small quantity of hydrazine or aniline at the
beginning of the rocket operation. Liquids that ignite with air (zinc diethyl or alu-
minum triethyl), when preloaded in the fuel piping, can accomplish a hypergolic
ignition. The flow diagram of the RD 170 Russian rocket engine in Fig. 11-2
shows several cylindrical containers prefilled with a hypergolic liquid, one for
each of the high-pressure fuel supply lines; this hypergolic liquid is pushed out
(by the initial fuel) into the thrust chambers and into the prebumers to start their
ignitions.

In vehicles with multiple engines or thrust chambers it is required to start two
or more together. It is often difficult to get exactly simultaneous starts. Usually
the passage or manifold volumes of each thrust chamber and their respective
values are designed to be the same. The temperature of the initial propellant
fed to each thrust chamber and the lead time of the first quantity of propellant
entering into the chambers have to be controlled. This is needed, for example,
in two small thrusters when used to apply roll torques to a vehicle. It is also
one of the reasons why large space launch vehicles are not released from their
launch facility until there is assurance that all the thrust chambers are started and
operating.

8.7. RANDOM VARIABLE THRUST

Only some applications require a randomly variable thrust from an engine.
Examples are descend on a planet or the moon, combat aircraft or upper stage
of an anti-ballistic missile. One of the advantages of liquid propellant rocket
engines is that they can be designed to throttle or randomly vary its thrust over
a wide range during flight. There have been several approaches to achieve vari-
able or randomly reduced thrust. Essentially all the approaches to vary the thrust
involve the thrust chamber and therefore this section has been placed into the
thrust chamber chapter. One of the early schemes was to use multiple engines or

8.7. RANDOM VARIABLE THRUST 327

multiple thrust chambers on the same engine and then stop the operation of one
or more of them. The Reaction Motors 6,000C-4 engine had four thrust chambers
and they could be turned on or off individually; this gave a step change in thrust.

Today one can distinguish between two modes of operation and both depend
on reducing the propellant flow. The first is moderate throttling, typically over
a thrust range of two or three, and this can usually be accomplished without
signihcant design changes to the engine. In one application the thrust is throttled
during the ascent of a booster vehicle in order to prevent excessive aerodynamic
heating. The other mode, often called deep throttling, varies the thrust by a factor
of between 6 to 30 for specific engines. It applies, for example, to a planetary
landing rocket engine with controlled deceleration. To achieve this deep throttling,
the engine requires some special features and some of these are mentioned below.

The thrust is almost proportional to the propellant mass flow (Eq. 2-14) and
therefore reducing that flow will reduce the thrust. With moderate throttling it
can be achieved by simultaneously closing the main fuel valve and the main
oxidizer valve (with matching flow characteristics) or alternatively by slowing
down the rotary speed of the turbopumps (by reducing the gas flow to the turbine)
with a hydraulically matched fuel pump and oxidizer pump. At lower flow the
pressure drop across the injector and the injection velocity will be diminished,
the atomization and mixing of the propellants will be somewhat less efficient and
the combustion will be less complete, and the chamber pressure will be reduced.
This usually causes a decrease in the specific impulse at the lower thrust levels
of perhaps 1.5 to 9%, depending on the specific engine design. This method
of varying the thrust is used in a number of booster engines for space launch
vehicles.

With deeper throttling the oxidizer and fuel flows tend to oscillate and will
usually no longer be at the original mixture ratio, the propellant injection streams
or sprays and their impingement locations will wander, and the flow is likely to
become erratic. To prevent these phenomena the engines have some special fea-
tures as illustrated by the lunar landing rocket engine; this engine was throttled
by a factor of up to 10:1, was developed by the predecessor of Northrop Grum-
man, and flew in the late 1960s and 1970s. Its propellant flow control valves
each included a cavitating venturi with a movable tapered pintle that allowed to
vary the flow area of the venturi throat of the thrust chamber valve. This assured
a predetermined steady reduced flow of propellants at all thrust levels, and main-
tained the mixture ratio to be the same as the full thrust value. Furthermore the
injection of the propellants was through a pintle injector with two annular slots
of a moving sleeve and the width of these slots could be reduced by an actuator
built into the injector. The lower right sketch of Fig. 8-3 shows this feature. It
allows high injection velocities of the liquid propellants, giving good atomization
and pretty good combustion and a relatively small loss in performance at low
thrust. The highest thrust variation of 357 to 1.0 was with the sustainer engines of
the Fance surface-to-surface missile. Its engine, developed by Rocketdyne (today
Pratt & Whitney Rocketdyne), hrst flew in the late 1960s. The specific impulse
at low thrust was poor (more than 15% loss).

328 THRUST CHAMBERS

Another approach to achieve variable thrust is to also vary the nozzle throat
area. See Eq. 3-31. This scheme requires a moveahle tapered pintle in the main
nozzle throat area. The pintle has to he made of heat-tolerant material or has to he
regeneratively cooled and the pintle position is usually hydraulically controlled.
This will allow to maintain essentially constant chamber pressure at all thrust
levels. Several experimental liquid propellant engines and also solid propellant
motors with a random variable nozzle throat area have been built and tested. To
the best of the author’s knowledge, none have flown.

For small reaction control thrusters the average thrust is usually reduced
by pulsing. It is accomplished by controlling the number of cycles or pulses
(each has one short fixed-duration thrust pulse plus a short fixed-duration zero-
thrust pause), by modulating the duration of individual pulses (with short pauses
between pulses), or alternatively by lengthening the pause between pulses.

8.8. SAMPLE THRUST CHAMBER DESIGN ANALYSIS

This example shows how a thrust chamber is strongly influenced by the overall
vehicle system requirements or the mission parameters and the vehicle design.
As outlined in the design section of Chapter If and in the discussion of the
selection of propulsion systems in Chapter 19, each engine goes through a series
of rationalizations and requirements that define its key parameters and its design.
In this example we describe one way of how the thrust chamber parameters are
derived from the vehicle and engine requirements. The overall system require-
ments relate to the mission, its purpose, environment, trajectories, reusability,
reliability, and to restraints such as allowable engine mass, or maximum dimen-
sional envelope. We are listing some, but not all of the requirements. It shows
how theory is blended with experience to arrive at the initial choices of the design
parameters. This example could be different, if it would be done by a different
design team.

Example 8-2. Here we define the application as a new upper stage of an existing
multistage space launch vehicle that will propel a payload into deep space. This means
continuous firing (no restart or reuse), operating in the vacuum of space (high nozzle area
ratio), modest acceleration (not to exceed 5 go), reasonably low cost, moderately high
performance (specific impulse), and a thrust whose magnitude depends on the payloads,
the flight path, and acceleration limits. The desired mission velocity increase of the stage
is 3400 m/sec. The engine is attached to its own stage, which is subsequently disconnected
and dropped from the payload stage. The payload stage (3500 kg) consists of a payload of
1500 kg (for scientific instruments, power supply, or communications and flight control
equipment) and its own propulsion systems (including propellant) of 2000 kg (for trajec-
tory changes, station keeping, attitude control, or emergency maneuvers). There are two
geometric restraints: The vehicle has an outside diameter of 2.0 m, but when the structure,
conduits, certain equipment, thermal insulation, fittings, and assembly are considered, it

8.8. SAMPLE THRUST CHAMBER DESIGN ANALYSIS 329

really is only about 1.90 m. The restraint on the stage length of 4.50 m maximum will
affect the length of the thrust chamber. We can summarize the key requirements:

Application

Payload

Desired velocity increase Au
Maximum stage diameter
Maximum stage length
Maximum acceleration

Uppermost stage to an existing
multistage launch vehicle
3500 kg

3400 m/sec in gravity-free vacuum
1.90 m
4.50 m
5 go

Decisions on Basic Parameters. The following engine design decisions or
parameter selection should he made early in the design process:

Propellant combination
Chamber pressure
Nozzle area ratio

Feed system, using pumps or pressurized tanks
Thrust level

From a performance point of view, the best propellant combination would
be liquid oxygen with liquid hydrogen. Flowever, this bipropellant would have
a low average specific gravity (0.36). There is not enough volume in this upper
stage to allow sufficient propellant mass for attaining the desired vehicle velocity
increment Am. The lower stages of the existing launch vehicle use liquid oxygen
with RP-1 fuel with an average specific gravity of about 1.014, and the launch
pad is already equipped for supplying these. The new stage is limited in volume
and cross section. Because of these factors the propellant combination of liquid
oxygen and RP-1 (a type of kerosene) is selected. From Fig. 5-1 we see that
the theoretical specific impulse is between 280 and 300 sec, depending on the
mixture ratio and whether we use frozen or shifting chemical equilibrium in
the nozzle flow expansion. This figure also shows that the maximum value of
the characteristic velocity c* is reached at a mixture ratio of about 2.30, which
is a fuel-rich mixture. We select this mixture ratio. Its combustion temperature
is lower than the mixture ratios with higher values, and this should make the
cooling of the thrust chamber easier. We will see later that cooling may present
some problems. Based on universal experience, we select a value of 4 part way
(about 40%) between the values for frozen and shifting equilibrium, namely 292
sec at the standard chamber pressure of 1000 psi or 6.895 MPa, and a nozzle big
enough for expansion to sea level. From Fig. 5-1 and Table 5-5 we find the
molecular mass to be 23 kg/kg-mol and the specific heat ratio k to be about 1 .24.
Later we will correct this value of 4 from this standard reference condition to
the actual vacuum specific impulse of the thrust chamber with its actual nozzle
exit area ratio.

330 THRUST CHAMBERS

Next we will select a chamber pressure, a nozzle area ratio, and a feed system
concept. Historically there has been favorable experience with this propellant
combination at chamber pressures between 100 and 3400 psia with nozzle area
ratios up to about 40 with both gas generator cycles and staged combustion
cycles, giving proof that this is feasible. The following considerations enter into
this selection:

1. Higher chamber pressures allow a smaller thrust chamber and (for the
same nozzle exit pressure) a shorter nozzle cone with a smaller nozzle exit
diameter. The thrust chamber is small enough for a toroidal tank to be built
around it, and this conserves stage length. This not only saves vehicle space
but usually also some inert mass in the vehicle and the engine. Figure 8-22
shows the relative sizes of thrust chambers for three chamber pressures
and two nozzle area ratios (e of 100 and 300). The nozzle length and exit
diameter cannot exceed the values given in the requirements, which, as can
be seen, rules out low chamber pressure or a large nozzle. The dimensions
shown are calculated later in this analysis.

2. The heat transfer rate, which varies with the gas-film coefficient, is almost
proportional to the gas density, which is nearly proportional to the chamber
pressure, as shown by Eq. 8-20 or 8-22. On some prior thrust cham-
bers there have been problems with the formation of solid carbon layer or
deposits either inside the cooling jacket (increasing wall temperatures) or
on the inner walls of the combustion chamber (the solid can flake off and
cause burnout). This favors a lower chamber pressure.

Pi = 8.962 MPa p, = 4.826 MPa p^ = 0.689 MPa

(1300 psia) (700 psia) (100 psia)

FIGURE 8-22. Comparison of thrust chamber sizes for three chamber pressures and
two nozzle area ratios (100 and 300).

8.8. SAMPLE THRUST CHAMBER DESIGN ANALYSIS 331

3. Concern over leak-free seals for both static and dynamic seals increases
with chamber pressure, which in turn causes all feed pressures also to
increase.

4. A feed system using pressurized gas is feasible, but its inert masses of tanks
and engine are favorable only if the chamber pressure is very low, perhaps
around 100 psia or less. The tanks for propellants and pressurizing gas
become very heavy and the thrust chamber will be very large and exceed
the dimensional restraints mentioned above. We therefore cannot use this
feed system or very low chamber pressures.

5. If we use a pump feed system, the power needed to drive the pumps
increases directly with chamber pressure p\. In a gas generator engine
cycle this means a slightly reduced performance as the value of p\ goes
up. For a staged combustion cycle it means high pressures, particularly
high-pressure hot gas flexible piping, and a more complex, heavier, and
expensive engine. We therefore select a gas generator cycle (see Fig. 1-4)
at a low enough chamber pressure so that the thrust chamber (and the other
inert hardware) will just fit the geometrical constraints, and the engine inert
mass and the heat transfer will be reasonable.

For these reasons we pick a chamber pressure of 700 psia or 4.825 MPa and an
area ratio of 100. With further analysis we could have picked p\ more precisely;
it could be slightly lower. Next we correct the specific impulse to the operating
conditions using a ratio of thrust coefficients. We can use Eq. 3-30 or interpolate
between Figs. 3-7 and 3-8 for a value of k = 1.24. The reference or standard
condition (see Fig. 3-6) is for a pressure ratio pi/p$ of 1000/14.7 = 68, which
corresponds to an area ratio of about 8. Then (Cf)standard = 1-58. For the actual
high-altitude operation the pressure ratio is close to inhnity. The nozzle has an
area ratio of 100; we can determine the thrust coefficient by interpolating with
pressure at k = 1.24. The result is (Cf)vacuum = 1-90. The new ideal specific
impulse value for a chamber threshold of 700 psia and a nozzle area ratio of
100 is therefore 292 x (1.90/1.58) = 351.1 sec. In order to correct for losses
(divergence, boundary layer, incomplete combustion, some film cooling, etc.)
we use a correction factor of 0.96 giving a thrust chamber specific impulse of
337.1 sec. The engine uses a gas generator and this will reduce the engine specific
impulse further by a factor of approximately 0.98 or (/i)engine = 330.3 sec or an
effective exhaust velocity of 3237 m/sec.

Stage Masses and Thrust Level. An estimate of the stage masses will next
be made. We assume that the inert hardware (tanks, gas, generator, turbopumps,
etc.) is about 7% of the propellant mass, which is conservative when compared
to existing engines. In a full-fledged engine design this number would be verified
or corrected once an estimated but more detailed mass budget becomes available.
From Eq. 4-7

Au/v _ '^p + 0-07OTp + 3500

^ ~~mf~ 007^7+3500

3400/3237

332 THRUST CHAMBERS

Solve for nip — 7483 kg. The final and initial masses ntf and mo of the stage are
then 4023 kg and 11,510 kg, respectively.

The maximum thrust is limited by the maximum allowed acceleration of
5go- It is Tmax = tUQU = 11,510 X 5 X 9.8 = 564,400 N. This would become
a relatively large and heavy thrust chamber. Considerable saving in inert mass
can be obtained if a smaller thrust size (but longer firing duration) is chosen.
Since this same thrust chamber is going to be used for another mission where an
acceleration of somewhat less than 1.0 go is wanted, a thrust level of 50,000 N or
11,240 Ibf is chosen. The maximum acceleration of the stage occurs just before
cutoff; it is a = F/mf = 50,000/4023 = 12.4 m/sec^ or about 1.26 times the
acceleration of gravity. This fits the thrust requirements.

The following have now been determined:

Propellant

Mixture ratio (0/F)

Thrust

Chamber pressure
Nozzle area ratio
Specific impulse (engine)

Specific impulse (thrust chamber)

Engine cycle

Usable propellant mass

Estimated nozzle exit exhaust velocity

Liquid oxygen and liquid kerosene
(RP-1)

2.30 (engine)

50,000 N or 11,240 Ibf
700 psia or 4.826 MPa
100

330.3 sec
337.1 sec
Gas generator
7478 kg

3237 m/sec or 10,613 ft/sec

Propellant Flows and Dimensions of Thrust Chamber. From Eq. 2-6 we

obtain the propellant mass flow:

m=F/c = 50,000/3237 = 15.446 kg/sec

When this total flow and the overall mixture ratio are known, then the fuel flow
rhf and oxidizer flow tko for the engine, its gas generator, and its thrust chamber
can be determined from Eqs. 6-3 and 6-4 as shown below:

riif = OT/(r + 1) = 15.446/(2.3 + 1) = 4.680 kg/sec

tho = mr/{r + 1) = (15.446 x 2.30)/3.30 = 10.765 kg/sec

The gas generator flow rhgg consumes about 2.0% of the total flow and operates
at a fuel-rich mixture ratio of 0.055; this results in a gas temperature of about
890 K:

(^/)gg = 0.2928 kg/sec (^o)gg = 0.0161 kg/sec

The flows through the thrust chamber are equal to the total flow diminished by
the gas generator flow, which is roughly 98.0% of the total flow or 15.137 kg/sec:

(mf )tc = 4.387 kg/sec

(Wo)tc = 10.749 kg/sec

8.8. SAMPLE THRUST CHAMBER DESIGN ANALYSIS 333

The duration is the total effective propellant mass divided by the mass flow
rate:

tj, = nip/thp = 7478/15.446 = 484.1 sec or a little longer than 8 min

The nozzle throat area is determined from Eq. 3-31. This neglects start and stop
transients:

At = F/{PiCf) = 50,000/(4.826 x 10® x 1.90) = 0.005453 m^ or 54.53 cm^

The nozzle throat diameter is Z), = 8.326 cm. The internal diameter of the
nozzle at exit A 2 is determined from the area ratio of 100 to be D 2 = x/TOO x Dj
or 83.26 cm. A shortened or truncated bell nozzle (as discussed in Section 3.4)
will be used with 80% of the length of a 15° conical nozzle, but with the same
performance as a 15° cone. The nozzle length (from the throat to the exit) can
be determined by an accurate layout or by an equivalent 15° conical nozzle exit
L = (D 2 — Dt)/(2 tan 15) as 139.8 cm. For an 80% shortened bell nozzle this
length would be about 111.8 cm. The contour or shape of a shortened bell nozzle
can be approximated by a parabola (parabola equation is = 2px). Using an
analysis (similar to the analysis that resulted in Fig. 3-14), the maximum angle
of the diverging section at the inflection point would be about 6>, = 34° and the
nozzle exit angle 9e = 7°. The approximate contour consists of a short segment
of radius 0.4r( of a 34° included angle (between points T and I in Fig. 3-14) and
a parabola with two known points at / and E. Knowing the tangent angles (34
and 7°) and the y coordinates [y^ = r 2 and y, = r, + 0.382 r,(l — coi0,)] allows
the determination of the parabola by geometric analysis. Before detail design
is undertaken, a more accurate contour, using the method of characteristics, is
suggested.

The chamber diameter should be about twice the nozzle throat diameter to
avoid pressure losses in the combustion chamber {E>c = 16.64 cm). Using the
approximate length of prior successful smaller chambers and a characteristic
length L* of about 1.1m, the chamber length (together with the converging nozzle
section) is about 11.8 in. or 29.9 cm. The overall length of the thrust chamber
(169 cm) is the sum of the nozzle length (1 1 1.8 cm), chamber (29.9 cm), injector
thickness (estimated at 8 cm), mounted valves (estimated at 10 cm), a support
structure, and possibly also a gimbal joint. The middle sketch of the three thrust
chambers in Fig. 8-22 corresponds roughly to these numbers.

We have now the stage masses, propellant flows, nozzle, and chamber config-
uration. Since this example is aimed at a thrust chamber, data on other engine
components or parameters are given only if they relate directly to the thrust
chamber or its parameters.

Next we check if there is enough available vehicle volume (1.90 m diameter
and 4.50 m long) to allow making a larger nozzle area ratio and thus gain a little
more performance. First we determine how much of this volume is occupied
by propellant tanks and how much might be left over or be available for the

334 THRUST CHAMBERS

thrust chamber. This analysis would normally be done by tank design specialists.
The average density of the propellant mixture can be determined from Eq. 7-1
to be 1014 kg/m^ and the total usable propellant of 7478 kg. Using densities
from Table 7-1 the fuel volume and the oxidizer volume can be calculated to be
2.797 and 4.571 m^, respectively. For a diameter of 1.90 m, a nearly spherical
fuel tank, a separate oxidizer cylindrical tank with elliptical ends, 6% ullage, and
2% residual propellant, a layout would show an overall tank length of about 3.6
m in a space that is limited to 4.50 m. This would leave only 0.9 m for the length
of the thrust chamber, and this is not long enough. Therefore we would need to
resort to a more compact tank arrangement, such as using a common bulkhead
between the two tanks or building a toroidal tank around the engine. It is not
the aim to design the tanks in this example, but the conclusion affects the thrust
chamber. Since the available volume of the vehicle is limited, it is not a good
idea to try to make the thrust chamber bigger.

This diversion into the tank design shows how a vehicle parameter affects
the thrust chamber design. For example, if the tank design would turn out to be
difficult or the tanks would become too heavy, then one of these thrust chamber
options can be considered: (1) go to a higher chamber pressure (makes the thrust
chamber and nozzle smaller, but heavier), (2) go to a lower thrust engine (will
be smaller and lighter), (3) store the nozzle of the upper-stage thrust chamber in
two pieces and assemble them during the flight once the lower stages have been
used and discarded (see extendible nozzle in Fig. 8-17; it is more complex and
somewhat heavier), or (4) use more than one thrust chamber in the engine (will
be heavier, but shorter and can provide roll control). We will not pursue these or
other options here.

Heat Transfer. The particular computer program for estimating heat transfer
and cooling parameters of thrust chambers will depend on the background and
experience of specific engineers and rocket organizations. Typical computer pro-
grams divide the internal wall surface of the chamber and nozzle into incremental
axial steps. Usually in a preliminary analysis the heat transfer is estimated only
for critical locations such as for the throat and perhaps the chamber.

From Fig. 5-1 and Eq. 3-12 or 3-22 we determine the following gas temper-
atures for the chamber, nozzle throat region, and a location in the diverging exit
section. They are: T\ — 3600 K, T, = 3214 K, and = 1430 K at an area ratio
of 6.0 in the diverging nozzle section. The chamber and nozzle down to an exit
area ratio of 6 will have to be cooled by fuel. For this propellant combination and
for the elevated wall temperatures a stainless steel has been successfully used for
the inner wall material.

Notice that beyond this area ratio of about 6, the nozzle free-stream gas tem-
peratures are relatively low. Uncooled high-temperature metals can be used here
in this outer nozzle region. Radiation cooling, using a material such as niobium
(coated to prevent excessive oxidation) or carbon fibers in a nonporous carbon
matrix, is suitable between an area ratio of 6 and about 25. For the final large
nozzle exit section, where the temperatures are even lower, a lower cost material

8.8. SAMPLE THRUST CHAMBER DESIGN ANALYSIS 335

such as stainless steel or titanium is suggested. Ablative materials have been ruled
out because of the long duration and the aggressive ingredients in the exhaust
gas. The gas compositions of Figs. 5-2 and 5-3 indicate that some free oxygen
and hydroxyl are present.

We now have identified the likely materials for key chamber components. The
best way to cool the radiation-cooled exit segment of the nozzle (beyond area
ratio of 6) is to let it stick out of the vehicle structure; the heat can then be freely
radiated to space. One way to accomplish this is to discard the vehicle structure
around the nozzle end after stage separation and before second-stage start.

As in Fig. 8-8, the maximum heat transfer rate will be at the nozzle throat
region. A variety of heat transfer analysis programs are available for estimat-
ing this heat transfer. If a suitable computer program is not available, then an
approximate steady-state heat transfer analysis can be made using Eqs. 8-14 to
8-18 and the physical properties (specific heat, thermal conductivity, and den-
sity) of RP-1 at elevated temperatures. The film coefficients of Eqs. 8-22 and
8-24 are also needed. This is not done in this example, in part because data
tables for the physical properties would take up a lot of space and results are
not always reliable. Data from prior thrust chambers with the same propellants
indicate a heat transfer rate at the nozzle throat region exceeding 10 Btu/in.^-sec
or 1.63 X 10^ W/m^.

The RP-1 fuel is an unusual coolant since it does not have a distinct boiling
point. Its composition is not consistent and depends on the oil stock from which
it was refined and the refining process. It is distilled or evaporated gradually
over a range of temperatures. The very hot wall can cause the RP-1 to locally
break down into carbon-rich material and to partially evaporate or gasify. As
long as the small vapor bubbles are recondensed when they are mixed with the
cooler portions of the coolant flow, a steady heat transfer process will occur. If
the heat transfer is high enough, then these bubbles will not be condensed, may
contain noncondensable gases, and the flow will contain substantial gas bubbles
and become unsteady, causing local overheating. The recondensing is aided by
high cooling passage velocities (more than 10 m/sec at the throat region) and by
turbulence in these passages. A coolant flow velocity of 15 m/sec is selected for
the nozzle throat region (adequate to prevent big gas bubbles), 7 m/sec for the
chamber region and somewhat less, 3 m/sec for the cooled nozzle exit segment.

The material for the cooling jacket will be stainless steel to resist the oxidation
and erosion of the fast moving, aggressive hot gas, which contains a small amount
of free oxygen and hydroxyl species. The forced cooling by fuel will assure that
the temperatures of this stainless steel are well below its softening temperature
of about 1050 K.

The construction of the cooling jacket can be tubular, as shown in Figs. 8-1
and 8-9, or it can consist of milled channels as shown in Figs. 8-2 and 8-15.
The cross section of each tube or cooling channel will be a minimum at the throat
region, gradually become larger, and be about two or more times as large at the
chamber and diverging nozzle regions. The wall thickness (on the hot gas side)
should be as small as possible to reduce the temperature drop across the wall

336 THRUST CHAMBERS

(which reduces the thermal stresses and allows a lower wall temperature) and to
minimize the yielding of the material that occurs due to thermal deformation and
pressure loads. Figure 8-11 shows this behavior, but for a thick wall. Practical
considerations such as manufacturability, the number of test firings before flight,
the deformation under pressure loads, the temperature gradient and dimensional
tolerances also enter into the selection of the wall thickness. A inner wall thick-
ness of 0.5 mm and a cooling velocity of 15 m/sec have been selected for the
throat region of the cooling jacket. Milled slots (rather than tubes) have been
selected for this thrust chamber.

The selection of the number of milled slots, their cross sections, and the wall
thickness is a function of the coolant mass flow, its pressure, wall stresses, wall
material, and the shape of the channel. Figure 8-23 and Table 8-6 describe
the channel width and height for different numbers of channels and different
locations. The fuel coolant flow is diminished by the gas generator fuel flow
(0.293 kg/sec) and is about 4.387 kg/sec. For this flow and a cooling velocity
of 15 m/sec in the throat region the cumulative cross-sectional area of all the
channels is only about 3.62 cm^. The cooling velocity is lower in the chamber
and nozzle regions and the cumulative channel flow area will be larger there.
The variables are the number of channels, the thickness of the hot wall, the rib
thickness between channels, the cooling velocity, the gas temperature, and the
location along the thrust chamber profile. The number of channels or tubes will
determine the shape of the cross section, ranging from deep and thin to almost
square. The effect of varying the number of channels or channel dimensions and
shape is shown in Table 8-6. The minimum inert mass of the cooling jacket and
a low friction loss occur, when the shape (which varies axially throughout the
jacket) is on the average close to a square. On the basis of analyses, as shown
in the table, a 150-channel design has been selected for giving favorable cross
section, reasonable dimensions for ease of fabrication, good cooling, and often
low thermal wall stresses.

Reinforcing bands have to be put on the outside of the tubes or channels to
hold the internal gas pressure during operation, to contain the coolant pressures,
which cause heated walls wanting to become round, and any surge pressures
during the start transient or arising from water hammer in the lines. We assume a
surge pressure of 50% above chamber pressure and a steel strength of 120,000 psi.
In the chamber the inside diameter is 16.7 cm (6.57 in.), the walls and channels

Hot gas side
of inner wall

Wall

thickness

FIGURE 8-23. Segment of cooling jacket with milled channels and an electroformed
outer wall.

8.8. SAMPLE THRUST CHAMBER DESIGN ANALYSIS 337

TABLE 8-6. Alternative Milled Channel Configurations for Fuel (Cooling) Flow of
4.387 kg/sec

Throat Section

Chamber Section

Wall thickness

0.05 cm

Wall thickness

0.06 cm

Rib thickness

0.08 cm

Rib thickness

0.08

Total flow area

3.653 cm^

Total flow area

7.827 cm2

Flow velocity

15 m/sec

Flow velocity

7.0 m/sec

Number of

Channel

Number of

Channels

Channel

Channels

Width, cm

Depth, cm

Channels

Width, cm

Depth, cm

0.257

0.177

100

0.193

0.189

100

0.456

0.171

120

0.145

0.210

120

0.367

0.179

140

0.113

0.231

140

0.303

0.184

150

0.100

0.243

150

0.277

0.188

160

0.092

0.247

160

0.255

0.192

180

0.070

0.289

180

0.218

0.196

each are 0.3 cm deep, and the pressure is 700 psia or 4.826 MPa. If one band
allows the reinforcing of a length of chamber of 3.0 in., the cross-sectional area
of that reinforcing band will be

A = pDLjila) = [700 x 1.5 x (6.57 -F 0.3) x 3]/(2 x 120,000) = 0.0902 in.^

If the bands were 1.0 in. wide, its thickness would be 0.09 in.; and if it were 3
in. wide, it would be 0.03 in. thick. Large nozzle exit sections have been observed
to experience flutter or cyclic deformation, and therefore some stiffening rings
may be needed near the exit.

The capacity of the fuel to absorb heat (Eq. 8-25) is approximately
CpWi/ AT = 0.5 X 4.81 x 200 = 278,000 J/sec. The maximum AT is estab-
lished by keeping the fuel well below its chemical decomposition point. This
calculated heat absorption is less than the heat transfer from the hot gases.
It is therefore necessary to reduce the gas temperature near the chamber and
nozzle walls or to increase the heat absorption. This can be accomplished by

(1) introducing film cooling by injection into the chamber just ahead of the
nozzle, by (2) modifying the injection patterns, so that a cooler, fuel-rich thick
internal boundary layer is formed, or (3) by allowing some nucleate boiling in
the throat region. The analysis of these three methods is not given here. Item

(2) , supplementary cooling, is selected because it is easy to design and build
and can be based on extensive data of prior favorable experience. However, it
causes a small loss of performance (up to about 1% in specific impulse). The
specific impulse of the engine is then 327 sec.

338 THRUST CHAMBERS

Injector Design. The injector pattern can be any one of the several types shown
in Figs. 8-3 and 8-4. For this propellant combination we have used both doublets
(like and unlike), and triplets in the United States, and the Russians have used
multiple hollow double posts with swirling or rotation of the flow in the outer
annulus. Based on good experience and demonstrated combustion stability with
similar designs, we select a doublet self-impinging-type stream pattern (Fig. 8-3)
and an injector structure design similar to Fig 8-5. The impinging streams form
fans of liquid propellant, which break up into droplets. Oxidizer and fuel fans
alternate radially. We could also use a platelet design, like Fig. 8-6.

The pressure drop across the injector is usually set at values between 15 and
25% of the chamber pressure, in part to obtain high injection velocities, which
aid in atomization and droplet breakup. In turn this leads to more complete
combustion (and thus better performance) and to stable combustion. We will
use 20% or 140 psi or 0.965 MPa for the injector pressure drop. There is a
small pressure loss in the injector passages. The injection velocities are found
from Eqs. 8-1 and 8-5. The equation is solved for the area A, which is the
cumulative cross-section area of all the injection holes of one of the propellants
in the injector face.

With rounded and clean injection hole entrances the discharge coefficient will
be about 0.80 as shown in Table 8-2. Solving for the cumulative injection hole
area for the fuel and the oxidizer flows gives 1.98 cm^ for the fuel and 4.098 cm^
for the oxidizer. A typical hole diameter in this size of injector would be about
0.5 to 2.5 mm. We will use a hole size of 1.5 mm for the fuel holes (with
90% of the fuel flow) and 2.00 mm for the oxidizer hole size, resulting in 65
doublets of oxidizer holes and 50 doublets of fuel. By using a slightly smaller fuel
injection hole diameter, we can match the number of 65 doublets as used with
the oxidizer holes. These injection doublets will be arranged on the injector face
in concentric patterns similar to Fig. 8-5. We may be able to obtain a slightly
higher performance by going to smaller hole sizes and a large number of fuel
and oxidizer holes. In addition there will be extra fuel holes on the periphery
of the injector face to help in providing the cooler boundary layer, which is
needed to reduce heat transfer. They will use 10% of the fuel flow and, for a
0.5-mm hole diameter, the number of holes will be about 100. To make a good
set of liquid fans, equal inclination angles of about 25° are used with the doublet
impingements. See Fig. 8-7.

Igniter Dimensions. A pyrotechnic (solid propellant) igniter will be used. It
has to have enough energy and run long enough to provide the pressure and
temperature in the thrust chamber for good ignition. Its largest diameter has to
be small enough to be inserted through the throat, namely 8.0 cm maximum
diameter and it can be 10 to 15 cm long. The actual igniter will, most likely, be
smaller than this.

PROBLEMS 339

Layout Drawings, Masses, Flows, and Pressure Drops. We now have
enough of the key design parameters of the selected thrust chamber, so a pre-
liminary layout drawing can be made. Before this can be done well, we will
need some analysis or estimates on the thrust chamber manifolds for fuel and
oxidizer, valve mounting provisions and their locations, a nozzle closure during
storage, a thrust structure, and possibly an actuator and gimbal mount, if gim-
baling is required by the mission. A detailed layout or CAD (computer-aided
design) image (not shown in this analysis) would allow a more accurate picture
and a good determination of the mass of the thrust chamber and its center of
gravity both with and without propellants.

Estimates of gas pressures, liquid pressures (or pressure drops) in the flow
passages, injector, cooling jacket, and the valves are needed for the stress anal-
ysis, so that various wall thicknesses and component masses can be determined.
Material properties will need to be obtained from references or tests. A few of
these analyses and designs may actually change some of the data we selected or
estimated early in this sample analysis, and some of the calculated parameters
may have to be re-analyzed and revised. Further changes in the thrust chamber
design may become evident in the design of the engine, the tanks, or the inter-
face with the vehicle. The methods, processes, and fixtures for manufacturing
and testing (and the number and types of tests) will have to be evaluated and the
number of thrust chambers to be built has to be decided before we can arrive at
a reasonable manufacturing plan, a schedule, and cost estimates.

PROBLEMS

1. How much total heat per second can be absorbed in a thrust chamber with an inside
wall surface area of 0.200 ra? if the coolant is liquid hydrogen and the coolant
temperature does not exceed 145 K in the jacket? Coolant flow is 2 kg/sec. What is
the average heat transfer rate per second per unit area? Use the data from Table 7-1
and the following:

2. During a static test a certain steel thrust chamber is cooled by water in its cooling
jacket. The following data are given for the temperature range and pressure of the
coolant:

Heat of vaporization near boiling point
Thermal conductivity (gas at 21 K)

446 kJ/kg

0.013 W/m-K
0.128 W/m-K
0.165 W/m-K

(gas at 194.75 K)
(gas at 273.15 K)

Average water temperature
Thermal conductivity of water
Gas temperature
Specific gravity of water

100°F

1.07 X 10-“* Btu/sec-ftT°F/ft
4500°F
1.00

340 THRUST CHAMBERS

Viscosity of water
Specific heat of water
Cooling passage dimensions
Water flow through passage
Thickness of inner wall
Heat absorbed

Thermal conductivity of wall
material

2.5 X 10“^ Ibf-sec/ft^
1.3 Btu/lb-°F
i X i in.

0.585 Ib/sec
i in.

1.3 Btu/in.^-sec
26 Btu/hr-fP-°F/ft

Determine (a) the film coefficient of the coolant; (b) the wall temperature on the
coolant side; (c) the wall temperature on the gas side.

3. In the example of Problem 2 determine the water flow required to decrease the wall
temperature on the gas side by 100°F. What is the percentage increase in coolant
velocity? Assume that the various properties of the water and the average water
temperature do not change.

4 . Determine the absolute and relative reduction in wall temperatures and heat transfer
caused by applying insulation in a liquid-cooled rocket chamber with the following
data:

Tube wall thickness
Gas temperature
Gas-side wall temperature
Heat transfer rate
Liquid-film coefficient
Wall material

0.381 mm
2760 K
1260 K

15 MW/m^-sec
23 kW/m^-K

Stainless steel AISI type 302

A 0.2-mm-thick layer of insulating paint is applied on the gas side; the paint consists
mostly of magnesia particles. The average conductivity of this magnesia paint is
2.59W/m^-K/m over the temperature range. The stainless steel has an average thermal
conductivity of 140 Btu/hr-ft^-°F/in. and a specific gravity of 7.98

5. A small thruster has the following characteristics:

Propellants

Injection individual hole size
Injection hole pattern
Thrust chamber type

Specific gravities
Impingement point
Direction of jet momentum
r = 1.65 (fuel rich)

F = 300 Ibf
p\ = 250 psi
(Ap)inj = 50.0 psi

Nitrogen tetroxide and monomethyl
hydrazine

Between 0.063 and 0.030 in.

Unlike impinging doublet
Ablative liner with a carbon-carbon nozzle
throat insert

1.446 for oxidizer and 0.876 for fuel

0.25 in. from injector face

Parallel to chamber axis after impingement

(?5)actual — 251 sec

tt = 25 sec

Ai/A, = 3.0

{Cd)o = (Q)/ = 0.86

PROBLEMS 341

Determine the number of oxidizer and fuel injection holes and their angles. Make a
sketch to show the symmetric hole pattern and the feed passages in the injector. To
protect the wall, the outermost holes should all be fuel holes.

6. A large, uncooled, uninsulated, low-carbon-steel thrust chamber burned out in the
throat region during a test. The wall (0.375 in. thick) had melted and there were
several holes. The test engineer said that he estimated the heat transfer to have
been about 15 Btu/in.^. The chamber was repaired and you are responsible for the
next test. Someone suggested that a series of water hoses be hooked up to spray
plenty of water on the outside of the nozzle wall at the throat region during the
next test to prolong the bring duration. The steel’s melting point is estimated to be
2550°F. Because of the likely local variation in mixture ratio and possibly imperfect
impingement, you anticipate some local gas regions that are oxidizer rich and could
start the rapid oxidation of the steel. You therefore decide that 2150°F should be
the maximum allowable inner wall temperature. Besides knowing the steel weight
density (0.284 Ibf/in.^), you have the following data for steel for the temperature
range from ambient to 2150°F: the specibc heat is 0.143 Btu/lbm-°F and the thermal
conductivity is 260 Btu/hr-ft‘-°F/in. Determine the approximate time for running
the next test (without burnout) both with and without the water sprays. Justify any
assumptions you make about the liquid-blm coefficient of the water flow. If the
water spray seems to be worth while (getting at least 10% more burning time), make
sketches with notes on how the mechanic should arrange for this water flow so it will
be most effective.

7 . The following conditions are given for a double-walled cooling jacket of a rocket
thrust chamber assembly:

Rated chamber pressure

210 psi

Rated jacket pressure

290 psi

Chamber diameter

16.5 in.

Nozzle throat diameter

5.0 in.

Nozzle throat gas pressure

112 psi

Average inner wall temperature at throat region

1100°F

Average inner wall temperature at chamber region

800°F

Cooling passage height at chamber and nozzle exit

0.375 in.

Cooling passage height at nozzle throat

0.250 in.

Nozzle exit gas pressure

14.7 psi.

Nozzle exit diameter

10 in.

Wall material

1020 carbon steel

Inner wall thickness

0.08 in.

Safety factor on yield strength

2.5

Cooling fluid

RP-1

Average thermal conductivity of steel

250 Btu/hr-ft^-°F/in.

Assume other parameters, if needed. Compute the outside diameters and the thickness
of the inner and outer walls at the chamber, at the throat, and at the nozzle exit.

8. Determine the hole sizes and the angle setting for a multiple-hole, doublet impinging
stream injector that uses alcohol and liquid oxygen as propellants. The resultant

342 THRUST CHAMBERS

momentum should be axial, and the angle between the oxygen and fuel jets (yo + yj)
should be 60°. Assume the following:

(Cd)„

0.87

Chamber pressure

300 psi

(Cd)f

0.91

Fuel pressure

400 psi

71 Ib/ft^

Oxygen pressure

380 psi

51 Ib/ft^

Number of jet pairs

1.20

Thrust

250 Ibf

Actual specihc impulse 218 sec

Answers: 0.0197 in.; 0.0214 in.; 32.3°; 27.7°.

9 . Table 11-6 shows that the RD-120 rocket engine can operate down to 85% of full
thrust and at a mixture ratio variation of ±10.0%. In a particular static test the average
thrust was held at 96% of nominal and the average mixture ratio was 2.0% fuel rich.
Assume a 1.0% residual propellant, but neglect other propellant budget allowances.
What percentage of the fuel and oxidizer that have been loaded will remain unused
at thrust termination? If we want to correct the mixture ratio in the last 20.0% of the
test duration and use up all the available propellant, what would be the mixture ratio
and propellant flows for this last period?

10 . Make a simple cross-section sketch approximately to scale of the thrust chamber that
was analyzed in Section 8.8. The various dimensions should be close, but need not be
accurate. Include or make separate detailed-section sketches of the cooling jacket and
the injector. Also compile a table of all the key characteristics, similar to Table 8-1,
but include gas generator flows and key materials. Make estimates or assumptions
for any key data that is not mentioned in Section 8.8.

SYMBOLS

A area, rc? (ft^)

Afl projected area of linear aerospike ramp, (ft^)

Cp specific heat at constant pressure, J/kg-K (Btu/lbm°R)

c average liquid specific heat, J/kg-K (Btu/lbm°R)

Cci discharge coefficient

Cp specific heat at constant pressure J/kg-K (Btu/lbm R)

D diameter, m (ft)

E modulus of elasticity, N/m^ (Ibf/in.^), or radiation energy, kg-m^/sec^

/ friction loss coefficient, or geometric factor in radiation or final

(velocity)

go sea-level acceleration of gravity, 9.806 m/sec^ (32.17 ft/sec^)
h film coefficient, W/(m^-K); (Btu/(ft^-°R)

Is specific impulse, sec

k specific heat ratio

L length, m (ft)

L* characteristic chamber length, m (ft)

m mass, kg

SYMBOLS 343

m mass flow rate, kg/sec (Ib/sec)

p pressure, N/m^ or Pa (Ibf/in.^)

Pr Prandtl number (Cp/r/A:)

q heat transfer rate or heat flow per unit area, J/m^-sec (Btu/ft^-sec)

Q volume flow rate, m^/sec (ft^/sec), or heat flow rate, J/sec

R Reynold’s number Dvp/ p.

r flow mixture ratio (oxidizer to fuel); or radius, m (ft)

^ stress N/m^ (Ibf/in.^)

t time, sec, or thickness, m (ft)

ts stay time, sec

tw wall thickness, m (in.)

T absolute temperature, K (°R)

Am flight velocity increment m/sec (If/sec)

V velocity, m/sec (ft/sec)

Vi specific volume, m^/kg (ft^/lb)

Vc combustion chamber volume (volume up to throat), m^(ft^)

X, y coordinates of a parabola with constant p

Greek Letters

Yo angle between chamber axis and oxidizer stream

Yf angle between chamber axis and fuel stream

A finite differential

5 angle between chamber axis and the resultant stream

e nozzle exit area ratio (e = A 2 IA,)

e emmissivity of radiating surface, dimension gas less

6 angle

K thermal conductivity, J/(m^-sec-K)/m (Btu/in.^-sec^-°R/in.)

X coefficient of thermal expansion, m/m-K (in./in. -°R)

ji absolute gas viscosity, kg/(m-sec) or Ibf/ft^-sec

V Poisson ratio

p density, kg/m^(lbf/ft^)

a Stefan-Boltzmann constant (5.67 x 10“* W/m^-K"^); also stress

N/m2(lbf/in2) N/m2(lbf/in2)

Subscripts

am arithmetic mean

c chamber

/ fuel or final condition

g gas

gg gas generator

344

THRUST CHAMBERS

inj

2
3

injector

liquid

oxidizer

throat

thrust chamber
wall

wall on side of gas

wall on side of liquid

initial condition

inlet or chamber condition

nozzle exit condition

atmosphere or ambient condition

REFERENCES

8-1. G. P. Sutton, History of Liquid Propellant Rocket Engines, AIAA, Reston, VA,

8-2. V. Yang, M. Habiballah, J. Hulka, and M. Popp, (Eds.), Liquid Rocket Thrust
Chambers: Aspects of Modeling, Analysis, and Design, Progress in Astronautics
and Aeronautics (Series), Vol. 200, AIAA, Reston, VA, 2004.

8-3. R. D. McKown, “Brazing the SSME,” Threshold, an Engineering Journal of Power
Technology , No. 1, Rocketdyne Division of Rockwell International (now Pratt &
Whitney Rocketdyne.), Canoga Park, CA, March 1987, pp. 8-13.

8-4. R. A. Ellis, I. C. Lee, E. M. Payne, M. Lacoste, A. Lacombe, and P. loyes,
“Testing of the RL lOB-2 Carbon-Carbon Nozzle Extension,” AIAA Conference
Paper 98-3363, July 1998.

8-5. M. Niino, A. Kumakawa, T. Hirano, K. Sumiyashi, and R. Watanabe, “Life Pre-
diction of CIP Formed Thrust Chambers,” Acta Astronautica, Vol. 13, Nos. 6-7,
1986, pp. 363-369 (fatigue life prediction).

8-6. J. S. Porowski, W. J. O’Donnell, M. L. Badlani, B. Kasraie, and H. J. Kasper,
“Simplified Design and Life Predictions of Rocket Thrustchambers,” Journal of
Spacecraft and Rockets , Vol. 22, No. 2, March-April 1985, pp. 181-187.

8-7. J. S. Kinkaid, “Aerospike Evolution,” Threshold, The Boeing Co., Rocketdyne
Propulsion and Power, No. 18, Spring 2000, pp. 4-13.

8-8. T. T. Bui, J. E. Murray, C. E. Rodgers, S. Bartel, A. Cesaroni, and M. Dennet,
“Flight Research of an Aerospike Nozzle Using High Power Solid Rockets,” AIAA
paper 2005-3797, 2005.

8-9. A. J. Fortini and R. H. Tuffias, “Advanced Materials for Chemical Propulsion:
Oxide-Iridium/Rhenium Combustion Chambers,” AIAA Paper 99-2894, June 1999.

8-10. F. P. Incropera, D. P. DeWitt, T. L. Bergman, and A. S. Lavine, Introduction to
Heat Transfer, 5th d., John Wiley & Sons, Hoboken, NJ, 2006.

8-11. A. A. Samarskii and P. N. Vabishchevich, Computational Heat Transfer, Vol. 1.
Mathematical Modeling and Vol. 2. The Finite Difference Methodology, John
Wiley & Sons, New York, 1995 and 1996.

2006.

REFERENCES 345

8-12. R. W. Lewis, Pemmal Nithiarasu, and Kankanhalli Seetharamu, Fundamentals of
the Finite Element Method for Heat and Fluid Flow , John Wiley & Sons, Hoboken,
NJ, 2004.

8-13. D. R. Bartz, “Survey of Relationships between Theory and Experiment for Convec-
tive Heat Transfer in Rocket Combustion Gases,” in Advances in Rocket Propul-
sion, S. S. Penner (Ed.), AGARD, Technivision Services, Manchester, UK, 1968.

8-14. E. Mayer, “Analysis of Pressure Feasibility Limits in Regenerative Cooling of
Combustion Chambers for Large Thrust Rockets,” in Liquid Rockets and Propel-
lants, L. E. Bollinger, M. Goldsmith, and A. W. Lemmon, Jr. (Eds.), Academic
Press, New York, 1969, pp. 543-561.

8-15. J. M. Fowler and C. F. Warner, “Measurements of the Heat-Transfer Coefficients
for Hydrogen Flowing in a Heated Tube,” American Rocket Society Journal, Vol.
30, No. 3, March 1960, pp. 266-267.

8-16. P. A. Baudart, V. Duthoit, T. Delaporte, and E. Znaty, “Numerical Modeling of
the HM 7 B Main Chamber Ignition,” AIAA Paper 89-2397, 1989.

8-17. A. R. Casillas, J. Eninger, G. Josephs, J. Kenney, and M. Trinidad, “Control of
Propellant Lead/Lag to the LEA in the AXAF Propulsion System,” AIAA Paper
98-3204, July 1998.

CHAPTER 9

LIQUID PROPELLANT COMBUSTION
AND ITS STABILITY

In this chapter we treat the complex phenomena of the combustion processes in
the combustion chamber of a liquid bipropellant thrust chamber. We describe
in general terms the combustion behavior, the progress in analysis of combus-
tion, the several types of combustion instability with its undesirable effects, and
semiempirical remedies which avoid these. The objective is to operate at very
high combustion efficiencies and to prevent 100% of all the occurrence of disrup-
tive or destructive combustion instability. Thrust chambers should operate with
stable combustion over their full range of operating conditions. For a treatment
of these subjects see Refs. 9-1 to 9-7.

The combustion of liquid propellants is very efficient in well-designed thrust
chambers, precombustion chambers, or gas generators. Efficiencies of 95 to
99.5% are typical compared to turbojets or furnaces, which can range from
50 to 97%. This is due to the very high reaction rates at the high combustion
temperatures and the thorough mixing of fuel and oxidizer reaction species by
means of good injection flow distribution and gas turbulence. The losses are
largely due to incomplete burning or inadequate mixing (nonuniform mixing
ratio). For very small bipropellant thrust chambers or small gas generators,
where the injector has very few injection orihces or elements, the combustion
efficiency can be well below 95%.

9.1. COMBUSTION PROCESS

In describing the combustion processes, it is convenient and helpful to the
understanding to divide the combustion chamber into a series of discrete zones,
as shown in Fig. 9-1 for a typical configuration. It has a flat injector face

346

9.1 . COMBUSTION PROCESS 347

Injection/ Rapid- Streamtube

atomization combustion combustion Transonic-flow zone

Supersonic

expansion

zone

FIGURE 9-1. Division of combustion chamber into zones for analysis. (Modified from
Y. M. Advanced Chemical Rocket Propulsion, Academic Press, New York, 1987.)

with many small injection orifices for introducing both fuel and oxidizer liquids
as many discrete individual streams, jets, or thin sprays or sheets. The relative
thicknesses of these zones in the axial direction, their behavior, and their
transitions are influenced by the specific propellant combination, the operating
conditions (pressure, mixture ratio, etc.), the design of the injector, and chamber
geometry. The boundaries between the zones shown in Fig. 9-1 are really not flat
surfaces and do not display steady flow. They are undulating, dynamically mov-
able, irregular boundaries with localized changes in velocity, temporary bulges,
locally intense radiation emissions, or variable temperature. Table 9-1 shows
the major interacting physical and chemical processes that occur in the chamber.
This table is a modification of tables and data in Refs. 9-2, 9-3, and 9-7.

The combustion behavior is propellant dependent. If the cryogenic fuel hydro-
gen has been used to cool the thrust chamber, the hydrogen would be gaseous
and fairly warm (60 to 240 K); there would be no liquid hydrogen droplets and
no evaporation. With hypergolic propellants there is an initial chemical reaction
in the liquid phase when a droplet of fuel impinges on a droplet of oxidizer.
Experiments show that the contact can create local explosions and enough
energy release to suddenly vaporize a thin layer of the fuel and the oxidizer
locally at the droplet’s contact face; there immediately follows a vapor chemical
reaction and a blow-apart and breakup of the droplets. Occasionally there are
small local explosions and shock wave phenomena (Refs. 9-8 and 9-9).

The combustion process is usually analyzed as a steady-flow process, but in
reality it is not so. When observing any one location within the chamber, one
finds a turbulent flow with local fluctuations with time in pressure, flow, tem-
perature, density, mixture ratio, or radiation emissions. The boundaries between
zones identified in Fig. 9-1 or Table 9-1 are not in a single plane, do not have
steady flow, are more like an undulating changing surface, which locally moves

348 LIQUID PROPELLANT COMBUSTION AND ITS STABILITY

TABLE 9-1. Physical and Chemical Processes in the Combustion of Liquid Propellants
Injection Atomization Vaporization

Liquid jets or sprays enter
chamber at relatively low
velocities

Sometimes gas propellant is
injected

Partial evaporation of
liquids

Interaction of jets or sprays
with high-pressure gas

Impingement of jets or
sheets

Formation of liquid fans
or spray cones
Formation of droplets
Secondary breakup of
drops

Liquid mixing and some
liquid-liquid chemical
reaction

Oscillations of jets or fans
or spray liquid sheets as
they become unstable
during breakup
Vaporization begins and
some vapor reactions
occur

Droplet gasification and
diffusion

Further heat release from
local chemical reactions
Relatively low gas velocities
and some cross flow
Heat absorbed by radiation,
convection, and
conduction, from blowback
of turbulent gases from the
hot reaction zone
Acceleration to higher
velocities

Vaporization rate influenced
by turbulence, pressure or
temperature oscillations
and acoustic waves

Mixing and Reaction Expansion in Chamber

Turbulent mixing (three-dimensional)
Multiple chemical reactions and major
heat releases

Interactions of turbulence with droplets
and chemical reactions
Temperature rise reduces densities and
increases volume flow
Local mixture ratios, reaction rates, or
velocities are not uniform across
chamber and vary rapidly with time
Some tangential and radial flows, much of
it near the injector

Chemical kinetics causes attainment of
final combustion femperafure and final
equilibrium reacfion gas composifion
Gas dynamics displays turbulence and
increasing axial gas velocities
Formation of a boundary layer
Acceleration to high chamber velocities
Streamlined high-velocity axial flow
with very little cross flow
Flow toward nozzle
Heat transfer to walls

forward and backward. Some of the processes listed in the table occur sequen-
tially, while others occur simultaneously. Not all the listed processes happen
with each propellant. Three-dimensional turbulence happens in all parts of the
combustion chamber and nozzle. Most processes also create noise.

Rapid Combustion Zone

In this zone intensive and rapid chemical reactions occur at increasingly
higher temperature; any remaining liquid droplets are vaporized by convective
heating and gas pockets of fuel-rich and fuel-lean gases are mixed. The mixing
is accomplished by local three-dimensional turbulence and diffusion of the
gas species.

9.1 . COMBUSTION PROCESS 349

The further breakdown of the propellant chemicals into intermediate fractions
and smaller, simpler chemicals and the oxidation of fuel fractions occur rapidly
in this zone. The rate of heat release increases greatly, and this causes the specific
volume of the gas mixture to increase and the local axial velocity to increase by
a factor of 100 or more. The rapid expansion of the heated gases also forces a
series of small local transverse gas flows (cross flow) and even some localized
temporary back flow from hot high-buming-rate sites to colder low-burning-
rate sites. The liquid droplets that may still persist in the upstream portion of
this zone, do not follow the gas flow quickly, and are difficult to move in a
transverse direction. Therefore, zones of fuel-rich or oxidizer-rich gases will
persist according to the orifice spray pattern in the upstream injection zone.
The gas composition and mixture ratio across the chamber section become more
uniform as the gases move through this zone, but the mixture never becomes
truly uniform. As the reaction product gases are accelerated, they become hotter
(due to further heat releases) and the lateral velocities remain relatively small
compared to the increasing axial velocities. Our understanding of the process is
aided by high-speed photography See Ref. 9-7.

Injection/Atomization Zone

Two different liquids are injected, both with storable bipropellants and/or liquid
oxygen-hydrocarbon combinations. They are injected through orifices or slots at
velocities typically between 7 and 60 m/sec or about 20 to 200 ft/sec. The injector
design has a profound influence on the combustion behavior and some seemingly
minor design changes can have a major effect on instability. The pattern, sizes,
number, distribution, and types of orifices influence the combustion behavior, as
do the pressure drop, manifold geometry, or surface roughness in the injection
orifice walls. The individual jets, streams, or sheets break up into droplets by
impingement of one jet with another (or with a surface), by impingement of
conical spray sheets, by the inherent instabilities of liquid jets or sprays, or by
the interaction with gases at a different velocity and temperature. In this first
zone the liquids are atomized into a large number of small droplets (see Refs.
9-3, 9-6, 9-7, 9-9, and 9-10). Heat is transferred to the droplets by radiation
from the very hot rapid combustion zone and by convection from moderately hot
gases in the first zone. The droplets evaporate and create local regions rich either
in fuel vapor or oxidizer vapor.

This first zone is heterogeneous; it contains liquids and vaporized propellant
as well as some burning hot gases. With the liquid being located at discrete sites,
there are large gradients in all directions with respect to fuel and oxidizer mass
fluxes, mixture ratio, size and dispersion of droplets, or properties of the gaseous
medium. Chemical reactions begin to occur in this zone, but the rate of heat
generation is relatively low, in part because the liquids and the gases are still
relatively cold and in part because vaporization near the droplets causes fuel-rich
and fuel-lean regions which do not bum as quickly. Some hot gases from the
combustion zone are recirculated back from the rapid combustion zone, and they

350 LIQUID PROPELLANT COMBUSTION AND ITS STABILITY

can create local gas velocities that flow across the injector face. The hot gases,
which can flow in unsteady vortexes or turbulence patterns, are essential to the
initial evaporation of the liquids.

The injection, atomization, and vaporization processes are different if one of
the propellants is a gas. For example, this occurs in liquid oxygen with gaseous
hydrogen propellant in thrust chambers or precombustion chambers, where
liquid hydrogen has absorbed heat from cooling jackets and has been gasified.
Hydrogen gas has no droplets and does not evaporate. The gas usually has a
much higher injection velocity (above 120 m/sec) than the liquid propellant.
This causes shear forces to be imposed on the liquid jets, with more rapid
droplet formation and gasification. The preferred injector design for gaseous
hydrogen and liquid oxygen is different from the individual jet streams used
with storable propellants, as shown in Chapter 8.

Stream Tube Combustion Zone

In this zone oxidation reactions continue, but at a lower rate, and some addi-
tional heat is released. However, chemical reactions continue because the mixture
tends to be driven toward an equilibrium composition. Axial velocities are high
(200 to 600 m/sec), and the transverse convective flow velocities are relatively
small. Streamlines are formed and there is relatively little turbulent mixing across
streamline boundaries. Locally the flow velocity and the pressure fluctuate some-
what. The residence time in this zone is very short compared to the residence
time in the other two zones. The streamline type, inviscid flow, and the chemical
reactions toward achieving chemical equilibrium presist not only throughout the
remainder of the combustion chamber, but are also extended into the nozzle.

The residence time of the propellant material in the combustion chamber is
very short, usually less than 10 msec. Combustion in a liquid rocket engine is very
dynamic, with the volumetric heat release being approximately 370 MJ/m^-sec,
which is much higher than in turbojets. Further, the higher temperature in a rocket
causes chemical reaction rates to be several times faster (increasing exponentially
with temperature) than in turbojet.

9.2. ANALYSIS AND SIMULATION

For the purpose of analyzing the combustion process and its instabilities, it
has been convenient to divide the acoustical characteristics into linear and
nonlinear behavior. A number of computer simulations with linear analyses
have been developed over the last 45 years and have been used to understand
the combustion process with liquid propellant combustion devices and to predict
combustion oscillation frequencies. The nonlinear behavior (e.g., why does a dis-
turbance cause an apparently stable combustion to suddenly become unstable?)
is not well understood and not properly simulated. Mathematical simulations
require a number of assumptions and simplifications to permit feasible solutions

9.2. ANALYSIS AND SIMULATION 351

(see Refs. 9-1 to 9-5, 9-7, 9-11, and 9-12). Good models exist for relatively
simple phenomena such as droplets of a propellant vaporizing (see Refs. 9-8
and 9-10) and burning in a gaseous atmosphere or the steady-state flow of gases
with heat release from chemical reactions. The thermochemical equilibrium
principles mentioned in Chapter 5 also apply here. Some programs consider
turbulence and film cooling effects.

The following phenomena are usually ignored or greatly simplified: cross
flows; nonsymmetrical gradients; unsteadiness of the flow; time variations in the
local temperature, local velocity, or local gas composition; thermochemical reac-
tions at local off-design mixture ratios and at different kinetic rates; enhancement
of vaporization by acoustic fields (see Ref. 9-9); uncertainties in the spatial as
well as the size distribution of droplets from sprays; or drag forces on droplets.
It requires skilled, experienced personnel to use, interpret, and modify the more
complex programs so that meaningful results and conclusions can be obtained.
The outputs of these computer programs can give valuable help and confirmation
about the particular design and are useful guides in interpreting actual test results,
but by themselves they are not sufficient to determine the designs, select specific
injector patterns, or predict the occurrence of combustion instabilities.

All the existing computer programs known to the authors are suitable for
steady-state flow conditions, usually at a predetermined average mixture ratio and
chamber pressure. However, during the starting, thrust change (such as throttling),
and stopping transients, the mixture ratio and the pressure can change drastically.
The analysis of these transient conditions is more difficult.

The combustion is strongly influenced by the injector design. The following
are some of the injection parameters which influence combustion behavior:
injector spray or jet pattern, their impingement, hole sizes or hole distribu-
tion, droplet evaporation, injection pressure drop, mixture ratio, pressure or
temperature gradients near the injector, chamber/injector geometry, pressurant
gas saturation, initial propellant temperature, and liquid injection pressure drop.
Attempts to analyze these effects have met with only partial success.

Computational fluid dynamics (CFD) is a relatively new analytical tool
that can provide a comprehensive description of complex fluid dynamic and
thermodynamic behavior. It allows for a time history of all parameters and can
even include some nonlinear effects. Numerical approaches are used to evaluate
sets of equations and models that represent the behavior of the fluid. For
complex geometries the information has been tracked with more than 150,000
discrete locations and can include changes in gas composition, thermodynamic
conditions, equilibrium reactions, phase changes, viscous or nonviscous flow,
one-, two-, or three-dimensional flow and steady-state or transient conditions. It
has been applied to resonance cavities in injectors or chambers and to the flow
of burning gases through turbines. A comprehensive rocket combustion model
using CFD is not yet available, but could become useful in the future.

352 LIQUID PROPELLANT COMBUSTION AND ITS STABILITY

9.3. COMBUSTION INSTABILITY

If the process of rocket combustion is not controlled (by proper design), then
combustion instabilities can occur which can very quickly cause excessive pres-
sure vibration forces (which may break engine parts) or excessive heat transfer
(which may melt thrust chamber parts). The aim is to prevent all occurrences
of this instability and to maintain reliable operation (see Refs. 9-5 and 9-11 to
9-14). Much progress has been made in understanding and avoiding combustion
instability.

References 9-1 to 9-3 and 9-5 to 9-7 describe the analysis and behavior
of combustion instability in thrust chamberts as practiced in the United States.
References 9-13 and 9-14 present the same but as practiced in Russia and its
predecessor the Soviet Union. There are some differences in the assumptions,
analysis, and testing, but the basic approach is essentially similar. In both coun-
tries any new or modified thrust chamber and/or rocket engine has to be tested
to demonstrate that combustion instability will not occur.

Table 9-2 lists the principal types of combustion vibrations encountered in
liquid rocket thrust chambers (see Refs. 9-3 and 9-9). Admittedly, combustion
in a liquid rocket is never perfectly smooth; some fluctuations of pressure, tem-
perature, and velocity are always present. When these fluctuations interact with
the natural frequencies of the propellant feed system (with and without vehicle
structure) or the chamber acoustics, periodic superimposed oscillations, recog-
nized as instability, occur. In normal rocket practice smooth combustion occurs
when pressure fluctuations during steady operation do not exceed about ±5% of
the mean chamber pressure. Combustion that gives greater pressure fluctuations
at a chamber wall location which occur at completely random intervals is called
rough combustion. Unstable combustion, or combustion instability, displays orga-
nized oscillations occurring at well-defined intervals with a pressure peak that
may be maintained, may increase, or may die out. These periodic peaks, repre-
senting fairly large concentrations of vibratory energy, can be easily recognized
against the random-noise background (see Fig. 9-2).

Chugging, the first type of combustion instability listed in Table 9-2, stems
mostly from the elastic nature of the feed systems and structures of vehicles or
the imposition of propulsion forces upon the vehicle. Chugging of an engine
or thrust chamber assembly can occur in a test facility, especially with low cham-
ber pressure engines (100 to 500 psia), because of propellant pump cavitation,
gas entrapment in propellant flow, tank pressurization control fluctuations, and
vibration of engine supports and propellant lines. It can be caused by resonances
in the engine feed system (such as an oscillating bellows inducing a periodic
flow fluctuation) or a coupling of structural and feed system frequencies.

When both the vehicle structure and the propellant liquid in the feed system
have about the same natural frequency, then force coupling can occur, not
only to maintain but also to strongly amplify oscillations. Propellant flow rate
disturbances, usually at 10 to 50 Hz, give rise to low-frequency longitudinal
combustion instability, producing a longitudinal motion of vibration in the

9.3. COMBUSTION INSTABILITY 353

TABLE 9-2. Principal Types of Combustion Instability

Type and Word Description

Frequency
Range (Hz)

Cause Relationship

Low frequency, called chugging
or feed system instability

10-400

Linked with pressure interactions
between propellant feed system,
if not the entire vehicle, and
combustion chamber

Intermediate frequency, called
acoustic,® buzzing, or entropy
waves

400-1000

Linked with mechanical vibrations
of propulsion structure, injector
manifold, flow eddies,
fuel/oxidizer ratio fluctuations,
and propellant feed system
resonances

High frequency, called screaming,
screeching, or squealing

Above 1000

Linked with combustion process
forces (pressure waves) and
chamber acoustical resonance
properties

“Use of the word acoustic stems from the fact the frequency of the oscillations is related to com-
bustion chamber dimensions and velocity of sound in the combustion gas.

vehicle. This vehicle flight instability phenomenon has been called pogo
instability since it is similar to pogo jumping stick motion. Pogo instabilities
can occur in the large, long propellant feed lines of large vehcles such as space
launch vehicles or ballistic missiles. See Refs. 9-15 to 9-17.

Avoiding objectionable engine-vehicle coupled oscillation is best accom-
plished at the time of initial design of the vehicle, as contrasted to applying
“fixes” later as has been the case with rocket engines for the Thor, Atlas, and
Titan vehicles. Analytical methods exist for understanding the vibration modes
and damping tendencies of major vehicle components, including the propellant
tanks, tank pressurization systems, propellant flow lines, engines, and basic vehi-
cle structure. Figure 9-3, a simplified spring-mass model of a typical two-stage
vehicle, indicates the complexity of the analytical problem. Fortunately, the vibra-
tional characteristics of the assembly can be affected substantially by designing
damping (energy absorption) into the major components or subassemblies. Tech-
niques for damping pogo instability include the use of energy absorption devices
in fluid flow lines, perforated tank liners, special tank supports, and properly
designed engine, interstage, and payload support structures.

A partially gas-filled pogo accumulator has been an effective damping device;
it is attached to the main propellant feed line. Such an accumulator is used in
the oxidizer feed line of the Space Shuttle Main Engine (SSME) betwen the two
oxidizer turbopumps; it can be seen in Eigs. 6-1 and 6-12. The SSME fuel line
does not need such a damping device because the fuel has a relatively very low
density and a lower mass flow.

354 LIQUID PROPELLANT COMBUSTION AND ITS STABILITY

Smooth combustion

Time

FIGURE 9-2. Typical oscillograph traces of chamber pressure p\ with time for different
combustion events.

The dynamic characteristics of a propellant pump can also have an influence
on the pogo-type vibrations, as examined in Ref. 9-17. The pogo frequency will
change as propellant is consumed and the remaining mass of propellant in the
vehicle changes. The bending or flexing of pipes, joints or bellows, or long tanks
also has an influence.

Buzzing, the intermediate type of instability, seldom represents pressure
perturbations greater than 5% of the mean in the combustion chamber and
usually is not accompanied by large vibratory energy. It often is more noisy
and annoying than damaging, although the occurrence of buzzing may initiate
high-frequency instability. Often it is characteristic of coupling between the
combustion process and flow in a portion of the propellant feed system.
Initiation is thought to be from the combustion process. Acoustic resonance of
the combustion chamber with a critical portion of the propellant flow system,
sometimes originating in a pump, promotes continuation of the phenomenon.
This type of instability seems to be more prevalent in medium-size engines
(2000 to 250,000 N thrust or about 500 to 60,000 Ibf) than in large engines.

9.3. COMBUSTION INSTABILITY

355

mi -Payload

mj-Spacer, dome,
5^ barrel

ms-Oxidizer

m^-Dome, ^ barrel
m5- Between tanks
ms- Dome, barrel
m^-Fuei

mg- Dome, barrel
mg-Tail skirt

mi(j-Engine
mil -Forward skirt
mi2-Dome, H barrel

mi3- Oxidizer

m,4-Dome, \ barrel
between tanks
mij-Dome, barrel
^ between tanks

mis-Fuel

miy-Fuel cone,
barrel

mig-Pumps
mig-Tail skirt

A A

/ \ m,o- Engines

FIGURE 9-3. Typical two-stage vehicle spring-mass model used in analysis of pogo
vibration in the vertical direction.

356 LIQUID PROPELLANT COMBUSTION AND ITS STABILITY

The third type, screeching or screaming, has high frequency and is most
perplexing and most common in the development of new engines. Both liquid and
solid propellant rockets commonly experience high-frequency instability during
their development phase. Since energy content increases with frequency, this
type is the most destructive, capable of destroying an engine in much less than
1 sec. Once encountered, it is the type for which it is most difficult to prove that
the incorporated “fixes” or improvemenfs render the engine “stable” under all
launch and flight conditions. It can be treated as a phenomenon isolated to the
combustion chamber and not generally influenced by feed system or structure.

High-frequency instability occurs in at least two modes, longitudinal and
transverse. The longitudinal mode (sometimes called organ pipe mode) prop-
agates along axial planes of the combustion chamber, and the pressure waves
are reflected at the injector face and the converging nozzle cone. The trans-
verse modes propagate along planes perpendicular to the chamber axis and can
be broken down into tangential and radial modes. Transverse mode instability
predominates in large liquid rockets, particularly in the vicinity of the injec-
tor. Figure 9-4 shows the distribution of pressure at various time intervals in
a cylindrical combustion chamber (cross section) encountering transverse mode
instability. Two kinds of wave form have been observed for tangential vibrations.
One can be considered a standing wave that remains fixed in position while its
pressure amplitude fluctuates. The second is a spinning or traveling tangential
wave which has associated with it a rotation of the whole vibratory system. This
waveform can be visualized as one in which the amplitude remains constant
while the wave rotates. Combinations of transverse and longitudinal modes can
also occur and their frequency can also be estimated. See Refs. 9-3 to 9-6.

Energy that drives screeching is believed to be predominantly from
acoustically stimulated variations in droplet vaporization and/or mixing, local det-
onations, and acoustic changes in combustion rates. Thus, with favorable acoustic
properties, high-frequency combustion instability, once triggered, can rapidly
drive itself into a destructive mode. Invariable, a distinct boundary layer seems
to disappear and heat transfer rates increase by an order of magnitude, much
as with detonation, causing metal melting and wall bum-throughs, sometimes
within less than 1 sec. The tangential modes appear to be the most damaging,
heat transfer rates during instability often increasing 4 to 10 times. Often the
instantaneous pressure peaks are about twice as high as with stable operation.

One possible source of triggering high-frequency instability is a rocket
combustion phenomenon called popping. Popping is an undesirable random
high-amplitude pressure disturbance that occurs during steady-state operation of
a rocket engine with hypergolic propellants. It is a possible source for initiation
of high-frequency instability. “Pops” exhibit some of the characteristics of a
detonation wave. The rise time of the pressure is a few microseconds and the
pressure ratio across the wave can be as high as 7:1. The elimination of popping
is usually achieved by redesign of the injector rather than by the application of
baffles or absorbers. They are explained later in this chapter.

9.3. COMBUSTION INSTABILITY 357

(Initial condition)

(Half cycle later)

FIGURE 9-4. Simplified representation of transverse pressure oscillation modes at two
time intervals in a cylindrical combustion chamber. The solid line curves indicate pressures
greater than the normal or mean operating pressure and the dashed lines indicate lower
pressures. The N-N lines show the node locations for these wave modes.

Some combustion instabilities can be induced by pulsations in the liquid flow
originating in turbopumps. Unsteady liquid flow can be caused by irregular cavi-
tation at the leading edge of the inducer impellers or of the main pump impellers
(Ref. 9-18). Also, when an impeller’s trailing edge passes a rib or stationary vane
of the volute, a small pressure perturbation always occurs in the liquid flow that
travels downstream to the injector. These two types of pressure fluctuation can
be greatly amplified if they coincide with the natural frequency of combustion
vibrations in the chamber.

358 LIQUID PROPELLANT COMBUSTION AND ITS STABILITY

The estimated natural frequencies can be determined from the wavelength /,
or the distance traveled per cycle, and the acoustic velocity a (see Eq. 3-10).
The frequency, or number of cycles per second, is

Frequency = a/l = {\/l)^kTR' /^K (9-1)

where k is the specific heat ratio, R' the universal gas constant, DJI the estimated
molecular weight of the hot chamber gases, and T the local average absolute
temperature. The length of wave travel depends on the vibrational mode, as
shown in Fig. 9-4. Smaller chambers give higher frequencies.

Table 9-3 shows a list of estimated vibration frequencies for one version of
the Vulcain HM 60 rocket thrust chamber; it operates with liquid hydrogen and
liquid oxygen propellants at a vacuum thrust of 1008 kN, a nominal chamber
pressure of 10 MPa, and a nominal mixture ratio of 5.6 (see Ref. 9-19). The
data in the table are based on acoustic measurements at ambient conditions with
corrections for an appropriate sonic velocity correlation; since the chamber has
a shallow conical shape and no discrete converging nozzle section, the purely
longitudinal vibration modes would be weak; in fact, no pure longitudinal modes
were detected.

Figure 9-5 shows a series of time-sequenced diagrams of frequency-
pressure-amplitude measurements taken in the oxygen injector manifold of
the Vulcain HM 60 engine during the first 8 sec of a static thrust chamber
test while operating at off-nominal design conditions. Chugging can be seen
at low frequency (up to 500 Hz) during the first few seconds, and a natural
frequency around 1500 Hz is attributed to the natural resonance frequency of the
oxygen injector dome structure where the high-frequency pressure transducer
was mounted. The continued oscillations observed at about 500 and 600 Hz are
probably resonances associated with the feed system.

TABLE 9-3. Estimated Acoustic Hot Gas Frequencies for Nominal Chamber
Operating Conditions for the Vulcain HM-60 Thrust Chamber

Mode"

(L, T, R)

Frequency (Hz)

Mode"

(L, T, R)

Frequency (Hz)

(0, 1, 0)

2424

L1T3

(1, 3, 0)

6303

LlTl

(1, 1,0)

3579

(0, 4, 0)

6719

(0, 2, 0)

3856

L2R1

(2, 0, 1)

7088

(0, 0, 1)

4849

(0, 5, 0)

8035

L1T2

(1, 2, 0)

4987

TR21

(0, 2, 1)

8335

(0, 3, 0)

5264

(0, 0, 2)

8774

LlRl

(1,0, 1)

5934

Reprinted with AIAA permission from Ref. 9-19.

"Modes are classified as L (longitudinal), T (tangential), or R (radial), and the number refers to the
first, second, or third natural frequency.

9.3. COMBUSTION INSTABILITY 359

Frequency (Hz)

FIGURE 9-5. Graphical representation of a series of 40 superimposed frequency-
amplitude diagrams taken 0.200 sec apart during the start phase (for the first 8 sec) of
the Vulcain HM 60 thrust chamber. In this static hot-firing test the thrust chamber was
operating at 109 bar chamber pressure and an oxidizer-to-fuel mass flow mixture ratio of
6.6. (Used with permission from Ref. 9-19.)

Rating Techniques

Semiempirical techniques exist for artificially disturbing combustion in a rocket
thrust chamber during test operation and evaluating its resistance to instability
(see Refs. 9-6 and 9-20). These include: (1) nondirectional “bombs” placed
within the combustion chamber; (2) oriented explosive pulses from a “pulse
gun” directed through the chamber sidewall; and (3) directed flows of inert
gas through the sidewall into the chamber. Often heavy prototype thrust
chambers are used because they are less expensive and more resistant to damage
than flight-weight engines. Other techniques used less widely but which are
important, especially for small engines include: (1) momentary operation at
“off-mixture ratio;” (2) introduction of “slugs” of inert gas into a propellant
line; and (3) a purposeful “hard start” achieved by introducing a quantity of
unreacted propellant at the beginning of the operation.

The objective of these rating techniques is to measure and demonstrate the
ability of an engine system to return quickly to normal or rated operation and
stable combustion after the combustion process has intentionally been disturbed
or perturbed.

All techniques are intended to introduce shock waves into the combustion
chamber or to otherwise perturb the combustion process, affording opportunity
for measuring recovery time from a predetermined overpressure disturbance,
assuming stable combustion resumes. Important to the magnitude and
mode of the instability are the type of explosive charge selected, the size of
the charge, the location and direction of the charge, and the duration of the
exciting pulse. The bottom curve in Fig. 9-2 characterizes the recovery of
stable operation after a combustion chamber was “bombed.” The time interval
to recover and the magnitude of explosive or perturbation pressure are then
used to rate the resistance of the engine to instability.

360 LIQUID PROPELLANT COMBUSTION AND ITS STABILITY

The nondirectional bomb method and the explosive pulse-gun method are
the two techniques in more common use. The bomb that can be used in large
flight-weight thrust chambers without modification consists of six charges of
250 grains or 16.2 grams of explosive (PETN, RDX, or similar) encased in
a Teflon, nylon, or micarta case. Detonation of the bomb is achieved either
electrically or thermally. Although the pulse gun requires modihcation of
a combustion chamber, this technique affords directional control, which is
important to tangential modes of high-frequency instability and allows several
data points to be observed in a single test run by installing several pulse guns
on one combustion chamber. Charges most frequently used are 10, 15, 20, 40,
and 80 grains of pistol powder. Pulse guns can be fired in sequence, introducing
successive pressure perturbations (approximately 150 msec apart), each of
increasing intensity, into the combustion chamber.

Control of Instabilities

The control or really the complete elimination of instabilities is an important
task during the design and development of a rocket engine. The designer usually
relies on prior experience with similar engines and tests on new experimental
engines. Analytical tools are available with which to simulate and evaluate the
combustion process. The design selection has to be proven in actual experiments
to be free of instabilities over a wide range of transient and steady-state operating
conditions. Some of the experiments can be accomplished on a subscale rocket
thrust chamber that has a similar injector, but most tests have to be done on a
full-scale engine.

The design features to control instabilities are different for the three types
of vibrations described in Table 9-2. Chugging is usually avoided if there is no
resonance in the propellant feed system or its coupling with the elastic vehicle
structure. Increased injection pressure drop and the addition of artihcial damping
devices in the propellant feed lines have been used successfully. Chugging and
acoustical instabilities sometimes relate to the natural frequency of a particular
feed system component that is free to oscillate, such as a loop of piping that can
vibrate an injector dome, or a bellows whose oscillations cause a pumping effect.

With the choice of the propellant combination usually fixed early in the plan-
ning of a new engine, the designer can alter combustion feedback (depressing the
driving mechanism) by altering injector details (such as changing the injector hole
pattern, hole sizes, or by increasing the injection pressure drop), or alternatively
by increasing acoustical damping within the combustion chamber. Of the two
methods, the second has been favored in recent years because it is very effective,
it is better understood, and theory fits. This leads to the application of injector
face baffles, discrete acoustic energy absorption cavities, and combustion chamber
liners or changes in injector design, often by using a trial-and-error approach.

Injector face baffles (see Fig. 9-6) were a widely accepted design practice in
the 1960s for overcoming or preventing high-frequency instability. Baffle design
is predicated on the assumption that the most severe instability oscillations, along

9.3. COMBUSTION INSTABILITY 361

Ignition flame tube

Oxygen
from
oxygen valve

Spark

Fluted oxidUer posts where
hot hydrogen evaporates the oxygen

Fuel inlet from
hot gas manifold

Cold hydrogen cavity

Five compartment
baffle with 75 cooled
injection posts

Primary injection
plate (transpiration
cooled) with 525 main
injection elements

Oxygen inlet
manifolds

Thrust

load

transmitting
cone

FIGURE 9-6. Main injector assembly of the Space Shuttle Main Engine showing baffle
with five outer compartments. (Courtesy of Pratt & Whitney Rocketdyne.)

with the driving source, are located in or near the injector-atomization zone at the
injector end of the combustion chamber. The baffles minimize influential coupling
and amplification of gas dynamic forces within the chamber. Obviously, baffles
must be strong, have excellent resistance to combustion temperatures (they are
usually cooled by propellant), and must protrude into the chamber enough to be
effective, yet not so far as to act like an individual combustion chamber with
its own acoustical characteristics. The number of baffle compartments is usually
odd. An even number of compartments enhances the standing modes of insta-
bility, with the baffles acting as nodal lines separating regions of relatively high
and low pressure. The design and development of baffles remains largely empir-
ical. Generally, baffles are designed to minimize acoustical frequencies below
4000 Hz, since experience has shown damaging instability is rare at frequencies
above 4000 Hz.

Various mechanisms of energy absorption or vibration damping exist in a
thrust chamber. Damping by wall friction in combustion chambers is not sig-
nificant. The exhaust nozzle produces the main damping of longitudinal mode
oscillations; the reflection of waves from the convergent nozzle entrance departs
from that of an ideal closed end. The principal damping source affecting propa-
gation in the transverse plane is combustion itself. The great volumetric change
in going from liquid to burned gases and the momentum imparted to a particle
(solid or liquid) both constitute damping phenomena in that they take energy
from high instantaneous local pressures. Unfortunately, the combustion process
can generate a great deal more pressure oscillation energy than is absorbed by
its inherent damping mechanism.

362 LIQUID PROPELLANT COMBUSTION AND ITS STABILITY

Volume ^ Orifice

V , ' diameter

T d

// y/.

L + AL
Helmholtz resonator

Spring

1 Mass

y//////////////M

Friction resistance

Mechanical analogy
of acoustic cavity

Chamber \wall

two cavities cavities

FIGURE 9-7. Diagram of acoustic energy absorber cavities at the periphery of an injec-
tor. In this thrust chamber the cavity restriction is a slot (in the shape of sections of a
circular arc) and not a hole. Details of the chamber cooling channels, injector holes, or
internal feed passages are not shown.

Acoustical absorbers are applied usually as discrete cavities along or in the
wall of the combustion chamber near the injector end. Both act as a series of
Helmholtz resonators that remove energy from the vibratory system which other-
wise would maintain the pressure oscillations. Figure 9-7 shows the application
of discrete cavities (interrupted slots) at the “comer” of the injector face. The cor-
ner location usually minimizes the fabrication problems, and it is the one location
in a combustion chamber where a pressure antinode exists for all resonant modes
of vibration, including longitudinal, tangential, radial, and combinations of these.
Velocity oscillations are minimal at this point, which favors absorber effective-
ness. Transverse modes of instability are best damped by locating absorbers at
the corner location. Figure 9-7 also shows a Helmholtz resonator cavity and its
working principles in simple form. Taking one resonator element, the mass of
gas in the orifice with the volume of gas behind it forms an oscillatory system
analogous to the spring-mass system shown (see Ref. 9-21).

Absorption cavities designed as Helmholtz resonators placed in or near the
injector face offer relatively high absorption bandwidth and energy absorbed
per cycle. The Helmholtz resonator (an enclosed cavity with a small passage
entry) dissipates energy twice each cycle (jets are formed upon inflow and out-
flow). Modern design practice favors acoustic absorbers over baffles. The storable

9.3. COMBUSTION INSTABILITY 363

propellant rocket engine shown in Fig. 8-2 has acoustic absorption cavities in
the chamber wall at a location next to the injector.

The resonance frequency / of a Helmholtz cavity can be estimated as

Here a is the local acoustic velocity, A is the restrictor area, A = {7r/4)d^, and
other symbols are as shown in Fig. 9-7. The AL is an empirical factor between
0.05 and 0.9L to allow for additional oscillating gas mass. It varies with the
L/d ratio and the edge condition of the restricted orifice (sharp edge, rounded,
chamfered). Resonators in thrust chambers are tuned or designed to perform their
maximum damping at predicted frequencies.

Small changes in injector geometry or design can cause an unstable combustion
to become stable and vice versa. New injectors, therefore, use the design and
geometry of proven, stable prior designs with the same propellants. For example,
the individual pattern of concentric tube injector elements used with gaseous
hydrogen and liquid oxygen (shown in Fig. 8-4d) are likely to be more stable,
if the hydrogen gas is relatively warm and the injection velocity of the hydrogen
is at least 10 times larger than that of the liquid oxygen.

There are other combustion vibration remedies besides those that have
already been mentioned, such a baffles or resonance cavities. This includes
higher injection pressure drop, avoiding resonance with certain key engine
component structures, changing critical injector dimension details, changing or
modifying the propellant, or adding liners. References 9-3 and 9-6 describe
several others, such as cooled flame holders, eliminating the entry of gases
into the chamber during combustion, changing the locations of maximum heat
release, varying critical dimensions on the injector elements, varying the number
of injector elements and their distribution over the cross section of the chamber,
changing the mixture ratio to be nonuniform near the face of the injector, or
temporary consumable baffles, which are effective only during the start.

In summary, the designer needs to (I) use data from prior successful engines
and simulation programs to establish key design features and estimate the likely
resonances of the engine and its major components, (2) design the feed system
and structure to avoid these resonances, (3) use a robust injector design that will
provide good mixing and dispersion of propellants and be resistant to distur-
bances, and (4) if needed, include tuned damping devices (cavities) to overcome
acoustic oscillations. To validate that a particular thrust chamber is stable, it is
necessary to test it over the range of likely operating conditions without encoun-
tering instability. An analysis is needed to determine the maximum and minimum
likely propellant temperatures, maximum and minimum probable chamber pres-
sures, and the highest and lowest mixture ratios, using a propellant budget as
shown in Section 11.1. These limits then establish the variations of test condi-
tions for this test series. Because of our improved understanding, the amount of
testing needed to prove stability has been reduced compared to that required 20
to 40 years ago.

(9-2)

364 LIQUID PROPELLANT COMBUSTION AND ITS STABILITY

PROBLEMS

1 . For a particular liquid propellant thrust chamber the following data are given:

Assume the gas composition and temperature to be uniform in the cylindrical chamber
section. State any other assumptions that may be needed. Determine the approximate
resonance frequencies in the first longitudinal mode, radial mode, and tangential mode.

2. Explain how the three frequencies from Problem 1 will change with combustion tem-
perature, chamber pressure, chamber length, chamber diameter, and throat diameter.

3. Why does heat transfer increase during combustion instability?

4 . Prepare a list of steps for undertaking a series of tests to validate the stability of a new
pressure-fed liquid bipropellant rocket engine. State the assumptions.

5. Estimate the resonant frequency of a set of each of nine cavities similar to Fig. 9-7.
Here the chamber diameter D = 0.200 m, the slot width is 1.0 mm, and the width and
height of the cavity are each 20.0 mm. The walls separating the individual cavities are
10.0 mm thick. Assume L = 4.00 mm, AL = 2.00 mm, and a = 1050 m/sec.

REFERENCES

9-1. R. D. Sutton, W. S. Hines, and L. P. Combs, “Development and Application of a
Comprehensive Analysis of Liquid Rocket Combustion,” AIAA Journal, Vol. 10,
No. 2, February 1972, pp. 194-203.

9-2. K. K. Kuo, Principles of Combustion, 2nd ed., John Wiley & Sons, Hoboken, NJ,
2005.

9-3. V. Yang and W. Anderson (Eds.), Liquid Rocket Engine Combustion Instability,
Vol. 169 of Progress in Astronautics and Aeronautics , AIAA, 1995, in particular
Chapter 1, F.E.C. Culick and V. Yang, “Overview of Combustion Instabilities in
Liquid Propellant Rocket Engines.”

9-4. W. Sirigano, A. Merzhanov, and L. De Luca, Advances in Combustion Science:
In Honor of Ya. B. ZeVdovich, Progress in Astronautics and Aeronautics Series,
V-173, American Institute of Aeronautics and Astronautics, 1997.

9-5. D. T. Harrje (Ed.), “Liquid Propellant Rocket Combustion Instability,” NASA SP-
194, U.S. Government Printing Office, No. 3300-0450, 1972.

9-6. G. P. Sutton, History of Liquid Propellant Rocket Engines, AIAA, 2006, Chapter
4.10 “Combustion and Vibration.”

Chamber presure

Chamber shape

Internal chamber diameter

Length of cylindrical section

Nozzle convergent section angle

Throat diameter and radius of wall curvature

Injector face

Average chamber gas temperature
Average chamber gas molecular weight
Specific heat ratio

Cylindrical
0.270 m
0.500 m
45°

0.050 m
Plat
2800 K
20 kg/kg-mol
1.20

68 MPa

REFERENCES 365

9-7. V. Yang, M. Habiballah, J. Hulka and M. Popp (Eds.), Liquid Rocket Thrust Cham-
bers: Aspects of Modeling, Analysis and Design, Progress in Astronautics and
Aeronautics (server), Vol 200, American Institute of Aeronautics and Astronautics,
Reston, VA 2004.

9-8. B. R. Lawyer, “Photographic Observations of Reactive Stream Impingement,”
Journal of Spacecraft and Rockets, Vol. 17, No. 2, March- April 1980,
pp. I34-I39.

9-9. M. Tanaka and W. Daimon, “Explosion Phenomena from Contact of Hypergolic
Liquids,” Journal of Propulsion and Power, Vol. 1, No. 4, 1984, pp. 314-316.

9-10. R. I. Sujith, G. A. Waldherr, J. I. Jagoda, and B. T. Zinn, “Experimental Investiga-
tion of the Evaporation of Droplets in Axial Acoustic Fields,” Journal of Propulsion
and Power, AIAA, Vol. 16, No. 2, March-April 2000, pp. 278-285.

9-11. P. Y. Liang, R. J. Jensen, and Y. M. Chang, “Numerical Analysis of the SSME
Prebumer Injector Atomization and Combustion Process,” Journal of Propulsion
and Power, Vol. 3, No. 6, November -December 1987, pp. 508-513.

9-12. M. Habiballah, D. Lourme, and F. Pit, “PHEDRE — Numerical Model for Combus-
tion Stability Studies Applied to the Ariane Viking Engine,” Journal of Propulsion
and Power, Vol. 7, No. 3, May-June 1991, pp. 322-329.

9-13. M. L. Dranovsky (author), V. Yang, F. E. C. Culick, and D. G. Talley (Eds.), Com-
bustion Instability in Liquid Rocket Engines, Testing and Development; Practices
in Russia, Progress in Astronautics and Aeronautics, Vol. 221, AIAA, Reston, VA,
2007.

9-14. M. S. Natanzon and F. E. C. Culick, Combustion Instability, Progress in Astro-
nautics and Aeronautics, Vol. 222, AIAA, Reston, VA, 2008.

9-15. B. W. Oppenheim and S. Rubin, “Advanced Pogo Analysis for Liquid Rockets,”
Journal of Spacecraft and Rockets , Vol. 30, No. 3, May-June 1993, pp. 360-373.

9-16. G. About, F. Bouvert, C. Bonna, N. David, and J. C. Lemoine, “A New Approach
of POGO Phenomenon Three-Dimensional Studies on the Ariane 4 Launcher,”
Acta Astronautica , Vol. 15, Nos. 6 and 7, 1987, pp. 321-330.

9-17. K. W. Dotson, S. Rubin, and B. M. Sako, “Mission Specific Pogo Stability Analysis
with Correlated Pump Parameters”, Journal of Propulsion and Power, Vol. 21,
No. 4. July-August 2005, pp. 619-626.

9-18. T. Shimura and K. Kamijo, “Dynamic Response of the LE-5 Rocket Engine Liquid
Oxygen Pump,” Journal of Spacecraft and Rockets, Vol. 22, No. 7, March-April
1985, pp. 195-200.

9-19. E. Kirner, W. Oechslein, D. Thelemann, and D. Wolf, “Development Status of the
Vulcain (HM 60) Thrust Chamber” AIAA Paper 90; 2255, July 1990.

9-20. “Guidelines for Combustion Stability Specifications and Verification Procedures for
Liquid Propellant Rocket Engines,” CPIA Publication 655, Chemical Propulsion
Information Agency, John Hopkins University, January 1997.

9-21. T. L. Acker and C. E. Mitchell, “Combustion Zone-Acoustic Cavity Interac-
tions in Rocket Combustors,” Journal of Propulsion and Power, Vol. 10, No 2,
March-April 1994, pp. 235-243.

CHAPTER 10

TURBOPUMPS AND THEIR GAS
SUPPLIES

10.1. INTRODUCTION

This is the first revised edition of this book, which has a separate chapter on tur-
bopumps (abbreviated here as TP) and their gas supplies. A TP is a high-pressure
high-precision piece of rotating high-speed power machinery usually consisting
of a turbine driving one or two centrifugal propellant pumps. Its purpose is to
take propellants from the vehicle’s propellant tanks, raise their pressure, and
deliver these propellants into the high-pressure piping system. The pressurized
propellants are then fed to one or more thrust chambers, where they are burned
and form hot gases. A TP is a unique key component of all larger liquid propel-
lant rocket engines with a pumped feed system and its design, fabrication, and
functions are unlike any of the other rocket engine components. This chapter dis-
cusses the various common types of TPs, outlines one design approach, describes
major TP components, and mentions some of their principal design issues.

A well-designed TP delivers the intended propellant flows at the intended
pump discharge pressures and mixture ratio, must have good enough reliabil-
ity so that it will not malfunction or fail during the intended flight duration,
and many will run at the highest practical energy efficiency of the pumps and
turbines. Furthermore the TP should not stimulate or cause any significant vibra-
tion (in the engine or vehicle) or should not be adversely affected by externally
caused vibrations, will function well over all operating conditions (such as dif-
ferent initial propellant temperatures, a range of ambient temperatures, start and
stop transients, accelerations in the flight direction and other directions) at the

366

10.2. DESCRIPTIONS OF SEVERAL TURBOPUMPS 367

minimum practical inert TP mass. It is important that the pumps will not cavitate
because the cavitation bubbles can reduce the steady nominal propellant flow and
can cause combustion instability. Cavitation is discussed later in Section 10.5.
The bearings and seals need to be adequately cooled by small secondary flows
of propellant within the TP assembly to prevent their overheating and malfunc-
tioning. There should not be any unexpected leakage in the seals and secondary
flow passages within a TP; this means leaks between bipropellants, between tur-
bine stages (gas) or pump stages (liquid), between a propellant and hot gas, or
between propellant or hot gas and the outside.

The selection of a specific TP configuration, such as those seen later in this
chapter, will depend on the propellant combination, the desired flows and pump
discharge pressures, the engine cycle, the available suction pressure, the number
of units to be built and delivered, and other factors. The available proven heat-
resistant materials and their maximum working temperatures for turbine blades
or the maximum load capacity of a bearing can influence the design selection.
The best placement of a TP within a rocket engine is usually a compromise
between several design considerations. For example, its inlet and outlet flanges
should be placed to minimize the piping, access bearings and turbine blades
for inspection on reusable engines, mounting which allows thermal expansion
without high stresses, or the simplicity of the design (e.g., minimum number of TP
components) — all these will influence the choice of the TP design configuration.
References 10-1 to 10-5 give further information on TP selection and design.

10.2. DESCRIPTIONS OF SEVERAL TURBOPUMPS

Typical TPs can be seen in Figs. 10-1 to 10-3 and 6-10, and many key compo-
nents can be identified by their callouts. Most of the part names in these figures
will be discussed in subsequent paragraphs of this chapter. TPs are also seen as a
component of an engine in Figs. 6-1, 6-11, 6-12, and 8-17. Figure 6-2 gives
a diagram of various common feed system arrangements; many have one or more
TPs. A number of different arrangements of the key TP components for several
common TP assemblies are shown in Fig. 10-4. This includes different numbers
of pumps and turbines, different component arrangements on TP shafts, with or
without gear cases, or with and without booster pumps. All will be discussed
later in this chapter. Figures 10-1 and 10-2 are experimental TPs. They were
selected for this book because they clearly identify the key components, (see Ref.
10-6). Figure 10-1 shows a single-stage propellant pump; it has a screw-type
axial-flow inducer impeller ahead of the main impeller, is driven by a single-
stage axial-flow turbine. The hot combustion gases, which drive this turbine, are
burned in a separate gas generator (or a precombustion chamber or prebumer)
at a mixture ratio that gives gases between 900 and 1200 K; this is sufficiently
cool, so that the hot turbine hardware (blades, nozzles, manifolds, or disks) still
have sufficient strength without needing forced cooling. The gases are expanded
(accelerated) in an annular set of converging -diverging supersonic turbine inlet

368 TURBOPUMPS AND THEIR GAS SUPPLIES

Entry slots
to turbine
nozzles

Turbine gas
discharge flange

Typical nozzle

Turbine blades
Turbine disk
Shaft seal

Angular ball
bearing for
axial loads

Inducer

impeller

Pump diffuser
vanes

Hydrostatic
bearing

Internal
passages for
supplying propellant
to bearings

Pump volute with
changing cross
section

Shrouded
impeller

bearing

Nut holding
inducer to
shaft

Pump

suction

flange

Pump

housing

Gas inlet
manifold and
turbine housing

Pump discharge
flange

FIGURE 10-1. Cut-away view of an experimental turbopump demonstrator with a
single-stage liquid oxygen pump impeller, an inducer impeller, and a single-stage turbine
(one row of blades) on the same shaft. (Courtesy of Pratt & Whitney Rocketdyne.)

nozzles, which are usually cast into the cast turbine inlet housing. The gases then
enter a set of rotating blades, which are mounted on a rotating wheel or turbine
disk. The blades essentially remove the tangential energy of the gas flow. The
exhaust gas velocity exiting from the blades is relatively low and its direction
is essentially parallel to the shaft. The pump is driven by the turbine through a
shaft which is supported by two experimental hydrostatic bearing. The propellant
enters the pump through an inducer, a special impeller where the pressure of the
propellant is raised only slightly (perhaps 5 to 10% of the total pressure rise).
This is just enough pressure so that there will be no cavitation as the flow enters
the main pump impeller. Most of the kinetic energy given to the flow by the
pump impeller is converted into hydrostatic pressure in the diffusers (the diffuser
vanes are not clearly visible, since they are inclined) and/or volutes of the pump.
The two hydrostatic bearings support the shaft radially. All bearings and shaft
seals create heat as they run. They are cooled and lubricated by small flows of
propellant, which are supplied from the pump discharge through drilled passages.

10.2. DESCRIPTIONS OF SEVERAL TURBOPUMPS 369

Housing for filtered
bearing supply

Pump

inlet

Cast pump housing with integral
crossover passages (Inconel 718)

Pump outlet

One-piece titanium rotor with inducer, two
impellers, turbine, and bearing surfaces

Split hydrostatic bearing
housings (Incoloy 909)

Radial in-flow turbine

Cast turbine housing with
vaneless internal volute

Turbine discharge
flange

Turbine gas
inlet flange

FIGURE 10-2. Exploded view of an advanced high-speed, two-stage liquid hydrogen
fuel pump driven by a radial flow turbine. (Used with permission of Pratt & Whitney
Rocketdyne, a part of United Technologies Corp.; adapted from Ref 10-7.)

One bearing (near the pump) is very cold and the other is hot, since it is close
to the hot turbine. The angular ball bearing accepts the axial net loads from the
unbalanced hydrodynamic pressures around the shrouded impeller, the inducer,
and also the turbine blades or the turbine disk.

A novel, experimental, high-speed, compact, and light weight liquid hydrogen
turbopump is shown in Fig. 10-2 and in Ref. 10-6 and 10-7. It was intended
to be used with an upper-stage hydrogen -oxygen rocket engine with a thrust of
about 50,000 Ibf (22.4 kN). The unique single-piece titanium rotor turns nomi-
nally at 166,700 rpm, has two machined sets of pump vanes, a machined inducer
impeller, a set of machined radial inflow turbine blades, and radial as well as
axial hydrostatic bearing surfaces. A small filtered flow of hydrogen lubricates
the hydrostatic bearing surfaces. The cast pump housing has internal crossover
passages between stages. The unique radial inflow turbine (3.2 in. dia.) produces
about 5900 hp at an efficiency of 78%. The hydrogen pump impellers are only
3.0 in. diameter and produce a pump discharge pressure of about 4500 psi at a
fuel flow of 16 Ibm/sec and an efficiency of 67%. A high pump inlet pressure of
about 100 psi is needed to assure cavitation-free operation. The turbopump can
operate at about 50% flow (at 36% discharge pressure and 58% of rated speed).
The number of pieces to be assembled is greatly reduced, compared to a more
conventional turbopump, thus enhancing its inherent reliability.

The geared turbopump in Fig. 10-3 has high turbine and pump efficiencies,
the rotary speed of the two-stage turbine is higher than the pump shaft speeds,
and the turbine is smaller than a comparable single-shaft TP. The auxiliary power

370 TURBOPUMPS AND THEIR GAS SUPPLIES

turbine

inlet
manifold

Oxygen
main impeller

Fuel

pump

impeller

Inducer

impeller

Fuel

outlet

Auxiliary

hydraulic

pump

2-stage spur
reduction gears

Volutes

FIGURE 10-3. Typical geared turbopump assembly similar to the one used on the RS-27
engine (Delta I and II launch vehicles) with liquid oxygen and RP-1 propellants. (Courtesy
of Praft & Whitney Rocketdyne.)

package (e.g., hydraulic pump) was used in early applications. The precision ball
bearings and seals on the turbine shaft can be seen, but the pump bearings and
seals are not visible in this figure.

Table 10-1 gives selected data on relatively large TPs for two large LOX/LH 2
rocket engines. It shows that the main LOX pumps have single-stage impellers,
while the main fuel pumps are multistage with two or three impellers in series.
The Space Shuttle Main Engine (SSME) has axial-flow inducer impellers on its
main pumps and also two booster pumps, which together raise the pressure of the
flows going to the inlets of the respective main centrifugal pumps. The booster
pump data is also listed. The Japanese LE-7 engine feed system does not use
booster pumps and it features pump inducers ahead of the main impellers. The
turbine blade shapes for both TPs are a combination of impulse turbines with
some reaction turbines contours. These terms will be explained later.

TABLE 10-1. Turbopump Characteristics

U §
Ij 00
2

U *0

0 X

1 O'

I J

^ C3

00 C
CO D

®0

(u O

S -o

'O • =

U "O

2 I +

(N ^

lO r- (N IT)
C 5 00 'O
<N

0^:

0. O

&. .2
N-1 X
<

'O- o o
o

o ^ 1 C“

in ^

Os (N r- <N

O O (N

s & 6

Q eS' Z

1 >.

c/2 CJ

2 c

a .2

1> cC
00 ^
S “

.C P.

a s

Q (2

00 Q fsj O

rn o
oo

Oi —

cn r-- (N

m o o o

O- 2
0) £

02 —

o^ p o o

o £

0:2 —

TO- O O o

o m -

cn VO ^

(N in

2 TD

■?. ^

s o

^ ON
NO (N
<N

in (N
in NO

O p o
NO f^r ON
<N

in ^

ON o o
NO (N (N

o —

# ^

cr _ O

a* s

c/2 Q.
flj flj
£ ^
2 15

l-i u.

O 3

H H

371

‘'Data courtesy of Pratt & Whitney Rocketdyne, at flight power level of 104.5% of design thrust.
*Data courtesy of Mitsubishi Heavy Industries, Ltd.

‘LPOTP, low-pressure oxidizer turbopump; HPFTP, high-pressure fuel turbopump.

Boost impeller stage for oxygen flow to prebumers.

372 TURBOPUMPS AND THEIR GAS SUPPLIES

10.3. SELECTION OF TURBOPUMP CONFIGURATION

The selection of the features, performance, and configuration of a specific new
TP arrangement can be complex. First of all it depends on the engine require-
ments for the particular flight application (thrust, propellant combination, mixture
ratio, chamber pressure, duration, low cost, schedule, etc.). These engine criteria
have to be evaluated, weighed, and prioritized. Furthermore it depends on the
engine cycle (see Section 6.6 and Fig. 6-9), propellant physical properties (such
as vapor pressure, density, or viscosity), the nominal flows and pump discharge
pressures (sum of chamber pressure, hydraulic losses in valves, cooling jacket,
injector, and pipes), minimum practical inert TP mass, high reliability (no flight
failures allowed), available suction pressure, maximum and minimum propel-
lant and hardware initial temperatures, and the components arrangement of the
assembly. It is also influenced by the size of the TP, the location of the TPs
inlet and outlet flanges relative to the thrust chamber inlets and the suction pipes
outlets, maximum allowable turbine gas inlet temperature (which depends on the
turbine material), number of hot starts during the life time, critical speed of the
rotating assembly, or the simplicity of the design (e.g., fewest number of parts).
If reuse is an engine requirement, then additional factors have to be considered,
such as ease of access to bearings, seals, or turbine blades for inspection of wear
or cracks. With throttling of the thrust the TP has to operate efficiently over a
range of speeds. There can be other TP selection criteria. Some of the preliminary
selection and design criteria are explained in Refs. 10-1 to 10-3 and 10-5 and
basic texts on pumps are in Refs. 10-8 and 10-9.

The shaft speed is related to the TP diameter or size and thus to the TP mass .
The tip speed of the pump impeller blades or the speed of the turbine blades at
their mean diameter can be the same at different diameters and rotational speeds.
Obviously, the highest practical shaft speed gives the smallest diameter, and thus
usually the lowest inert TP mass. This is often in conflict with other criteria, such
as cavitation avoidance, which is likely to be more difficult if the impeller speed
is increased. It will be discussed later in this section.

The arrangements of the various key TP components into an assembly can be
an important design criteria. Figure 10-4 shows several common arrangements.
The two most common are to have one turbine with two propellant pumps on the
same shaft (Figs. 10-4a and 10-4b) and two smaller separate TPs, one where
a turbine drives an oxidizer pump and one where another turbine drives a fuel
pump (Fig. 10-4e). In Fig. 10-4a the fuel pump inlet has a shaft going through
it and that will affect the suction pressure. In the TP in Fig. I0-4b the turbine
requires more seals. Although not discussed here, the placement and selection of
bearings and seals influences the TP selection.

Centrifugal pumps are used in all known production TPs, and these are basi-
cally constant-volume flow devices (Refs. 10-3, 10-4, 10-5, 10-8, and 10-9).
If the two propellants have similar densities (say within about 40%), such as
nitrogen tetroxide and UDMH or liquid oxygen and kerosene and the volume
flow of oxidizer and of fuel are similar, then the same type of impeller (running

1 0.3. SELECTION OE TURBOPUMP CONPIGURATION 373

L^-

Two pumps on same shaft
with outboard turbine.
Shaft goes through fuei
pump inlet.

Direct drive with turbine in
middle. Shaft goes through
turbine discharge manifold.

Two main pumps and two booster
pumps, each with its own gas turbine.

High-speed turbine and
fuel pump geared to
lower speed oxidizer
d pump.

Two turbines, each with one pump
Gas flow shown in parallel,
(alternate is gas flow in series,
first through one and then the
0 other turbine).

With gear case, turbine can
run faster. The two pumps
I have different speeds.

Two in-line shafts with flexible
g coupling C.

G G

Two main pumps and
two booster pumps, each
with own turbine.

FIGURE 10-4. Simplified diagrams of several different component arrangements of tur-
bopumps (TP); F, fuel pump; O, oxidizer pump; T, turbine; G, hot gas; C, shaft coupling;
and GC, gear case. (Used with permission from Ref. 10-1.)

at the same speed) can be used on both of them on a single shaft. If the densities
are quite different (e.g., liquid oxygen and liquid hydrogen), a single shaft would
not be practical or efficient. Instead two separate TPs are used — rotating at high
speed for the fuel pump and lower speed for the oxidizer pump — thus giving
good pump efficiencies. This corresponds to Fig. 10-4e.

It is at times difficult to package a relatively large TP into an engine package.
The two-TP configuration is often easier to integrate into an engine assembly,

374 TURBOPUMPS AND THEIR GAS SUPPLIES

both pumps will be able to run at a relatively high efficiency, the TP will have
fewer seals, and since the shafts are shorter, they are lighter for the same shaft
stiffness.

Figure 10-4g shows two shafts, in line with each other, with a mechanical cou-
pling between them (Ref. 10-10). In this example the fuel pump and the turbine
are on one shaft and the oxidizer pump on the other. The two shafts are connected
by a flexible coupling, and this allows considerable shaft misalignment, and as
a result some more generous fabrication and assembly tolerances in the TP. The
first two-shaft inline TP assembly was in the German V-2 rocket engine (1938)
with a six pin/sleeve type of flexible coupling. The Soviet Union developed a
curvic coupling serrated sleeve as a coupling and it was smaller and lighter. Ref-
erences 10-1 and 10-4 show a section view of these two TPs. The United States
did not produce a TP with two inline coupled shafts. When compared to a larger
single shaft, the smaller spans between bearings on each of the two shafts results
in substantially smaller shaft diameters, which in turn can reduce the mass of the
shafts, the diameter and mass of pumps and pump housings, and the inertia of
the rotary assembly and can allow some increases of the shaft speed.

Figures 10-4d and 10-4f show TPs with a gear case to transmit the power
from the turbine to the pumps (Ref. 10-11). The purpose is to allow the turbine
to run at high speed and the pumps to rotate at lower speeds, where the efficien-
cies are relatively high, thus minimizing the amount of gas generator propellant
needed to drive the TP. Gear cases for TPs were common in the United States
beginning in the early 1950s and one is still in use. In Fig. 10-4e the turbine
runs at a higher speed and the two pumps run at lower, but somewhat different
speeds. In Fig. 10-4f the fuel is liquid hydrogen; the fuel pump and the turbine
run at a higher speed on the same shaft and the oxygen pump is geared down to
a lower speed.

Figures 10-4c and 10-4h represent two arrangements for using booster TPs
in addition to the two main TPs. See Ref. 10-12. The purpose is to reduce
the vehicle’s tank pressure and inert mass further, thus improving the vehicle
performance. A typical booster TP provides about 10% of the propellant pressure
rise and the respective main pump about 90%. Booster pumps typically use axial-
flow impellers, and a few can operate with considerable cavitation in the leading
edges of the booster impellers. With a booster pump the pressure at the inlet to the
main pump is raised enough so cavitation does not occur in these higher pressure
impellers. The low rotating speed causes the booster pumps to have a relatively
large diameter. In some liquid propellant rocket engines they are larger than the
main pump. Since pressures in booster pumps are low, the walls are thin, and
the booster pump inert mass can be low. Most of the turbines of booster TPs are
driven by gas expansion, but some are powered by high-pressure liquid propellant
taken from the discharge of the main pump. The total number worldwide of rocket
engines with booster pumps is small, perhaps a dozen. Booster pumps provide
low suction pressure and good cavitation resistance to the main pumps of the
engines, but the engines are more complex and more costly.

Bearings and seals will be overloaded or fail if the shaft deflects (even by a
small amount like 0.001 in. in some TPs). See Refs. 10-13 and 10-14. In order to

1 0 . 3 . SELECTION OE TURBOPUMP CONPIGURATION 375

minimize deflection of the shaft, the shaft has to be stiff (less deflection for a given
side load) and that means large in diameter and heavy. The rotating assembly
(turbine, pumps, seals, bearings) has to be carefully balanced. An unbalance
causes a side load (which can be substantial at high rotational speed) and a
deflection of the shaft. Both accurate static and dynamic balances are usually
required during assembly. The operating speed of the rotating assembly is never
chosen to be the same as the resonant or critical shaft speed, where deflections
can become very high and the bearings, seals, or even the shaft can fail. The
synchronous or critical speed is usually well above the operating speed. A stiffer
shaft (larger diameter, heavier, a shorter distance between bearings, and less
overhang beyond the bearing) gives a higher critical speed and allows the pumps
to run at higher speed, which allows a lighter TP assembly. There are some TP
where the operating speed is actually above the critical speed. Here the shaft
momentarily goes trough the critical speed during startup and shutdown.

There are usually several secondary flow passages in a TP. A bearing, rotating
shaft seal, or impeller wear ring are all usually cooled and/or lubricated with a
very small flow of propellant, which is heated in the process and can then be
fed back to the suction side of a pump or dumped overboard. It requires a set
of small external tubes or internal passages in the TP assembly to route several
small flows for achieving this cooling and lubrication of rubbing surfaces. There
is a good number of seals and bearings that need a small propellant supply and
the distribution system for this propellant can be complex; some routes have to
be throttled to the desired small flow, and a mixing of unlike propellants has to
be avoided. The analysis of these secondary flow passages can be tedious. If a
bearing or seal is close to a turbine, they will get warm (needs more coolant) and
if they are pumping a cryogenic propellant, the adjacent hardware will become
very cold. The clearances of these bearings or seals have to be so designed that
the intended clearance is correct at the operating conditions.

Bubbles in the flow of propellant are indeed unwanted. They can be introduced
into the propellant feed pipes by extensive cavitation in the pumps, (see Refs.
10-4 and 10-9) by leaks in low-pressure suction piping, or by improper priming
or filling of the propellant lines before or during start. Bubbles reduce the effec-
tive density of the propellant, which in turn reduces the mass flow and the mixture
ratio and usually reduces the thrust level somewhat below the nominal value. If
the intended operating mixture ratio was fuel rich and bubbles were to reduce
the fuel flow, this mixture ratio will approach a stochiometric value; this creates
a very hot gas and would most likely cause a failure of the cooling jacket in less
than a second. A major concern with bubbles is that they can readily trigger com-
bustion instability (see Chapter 9), which often leads to thrust chamber failure.
The key design objectives are to select a high enough shaft speed (which gives
the smallest and lightest TP, but is still not prone to cavitation), a low enough
pump suction pressure (which gives the lightest vehicle tank mass, but this pro-
motes pump cavitation), and an impeller configuration, which will be strong and
efficient and prevent dangerous cavitation under all operating conditions.

A balance of axial fluid pressure forces within the TP rotating assembly and
its housings is necessary to prevent excessive axial hydraulic pressure forces

376 TURBOPUMPS AND THEIR GAS SUPPLIES

(on surfaces normal to the shaft) from overloading certain parts of the assembly
(Refs. 10-1 and 10-15). Unbalance would cause a small axial movement of the
rotating assembly (typically just a few thousands of an inch) and often intense
rubbing of stationary and rotating parts. These pressure forces are created by
the fluids flowing through the pump paths (increasing pressures) and the turbine
gases flowing through moving and stationary sets of turbine blades (decreasing
pressures), and by the radial placement of wear rings or seals. Some turbopumps
have balancing pistons and some use the fluid pressures on the back side of an
impeller or turbine disk to make an effective balance piston. Pressure on the
piston is controlled by a slight axial movement, which changes clearances at the
edge of the piston. During the transient operations (start, stop, change thrust),
there typically are brief periods of unbalance caused by changes in the pressure
distributions of the internal flows, but they are of short duration. Nevertheless
there can be some rubbing or axial contact of a rotating and a stationery surface
normal to the shaft during the brief periods of unbalance. If the axial rubbing
contact persists, there will be some damage to the TP.

10.4. FLOW, SHAFT SPEEDS, POWER, AND PRESSURE BALANCES

The design of a TP requires the careful balancing of the propellant flows, the
shaft speeds, and the power between the key pumps and turbines and the pressure
distribution of the propellant along its flow paths. For a TP with two pumps and a
gear transmission, the shaft speeds can be written as in Eq. 10-1. The relationship
between shaft speeds, torques, and power can readily be seen by reviewing a flow
diagram of a TP such as Fig. 1-4. With a gear case the relation between shaft
speeds can be written as

Here N is the shaft speed, the subscripts t, f, and o stand for turbine, fuel pump,
and oxidizer pump, respectively. The a„ and af are the gear ratios for the oxidizer
and fuel pumps. When there are no gears, then Ug and aj become 1.0. For a TP
similar to Fig. 10-4a and 10-4b:

Here Lh is the torque to overcome friction in bearings, seals, or auxiliaries. For
a TP as in Figs.l0-4a and 10-4b, the power of the turbine Pt has to equal the
sum of the powers of the pumps plus some losses. This power balance can also
be expressed as the product of the torque L and the speed N:

Nf = agNo = ajNf

( 10 - 1 )

Nt — Nf — Ng and L, — Lg -\- Lj -\- Lh

( 10 - 2 )

(10-3)

Pt = L,Nt = LfNf + LgNg + Pb

(10-4)

10.4. FLOW, SHAFT SPEEDS, POWER, AND PRESSURE BALANCES 377

Here Pt is the power for overcoming the friction in all the bearings and seals,
and in some TP conhgurations also the power loss in gears, and auxiliaries (such
as an oil pump for lubricating gears). The power used in bearing, seals, and
auxiliaries is usually small.

For the two separated TP scheme in Fig. 10-4e each of the TPs has its own
internal losses. It is obvious that the speeds N and the power P and even the
torques L must balance in this conhguration:

{N,)o = {Np)o and {N,)f = {Np)f (10-5)

{Pt)o = i.Pp)o + {Pb)c and = {Pp)f + (10-6)

The subscripts t and p stand for turbine and pump. The pressure balance
equation for the fuel line (the oxidizer line is the same) at a point just past the
fuel pump discharge flange can be written as

(P/)d = {Pf)s + (A/J/-)pump (10-7)

~ (Ap)i-nain fuel system T Pi
— ( Ap)gas generator fuel system T Pgg

Here the fuel pump discharge pressure (pf)d equals the fuel pump suction pres-
sure {App)s plus the pressure rise across the pump (Ap)pump- This discharge
pressure in turn has to equal the chamber pressure pi plus all the pressure drops
in the high-pressure main fuel flow system downstream of the pump and upstream
of the chamber. This usually includes the pressure losses in the cooling jacket,
the injector, the piping, and the open fuel valve. This pump discharge pressure
furthermore has to equal the gas generator combustion pressure pgg and all the
pressure losses between the fuel pump discharge and the gas generator combus-
tion chamber. A similar pressure balance is needed for the oxidizer systems.

The fourth type of balance is different from the other three (flow, pressure, and
power balance); it is aimed at preventing certain high-pressure axial hydraulic
loads and/or axial gas pressure loads from causing damage inside the TP assem-
bly See Refs. 10-1, 10-4, and 10-15. These axial internal loads (side forces
on impellers or turbine disks) can be very high, particularly in large sized TPs.
Certain ball bearings are capable of withstanding considerable axial loads, in
addition to the usual radial loads. The axial movement of the rotating assembly
is limited by design and is very small. This method of controlling axial forces
with a ball bearing is used in many TPs, but usually the maximum axial load
of ball bearings is limited; therefore it is used mostly with small-diameter TPs
or low-pressure TPs. A balance of axial fluid pressure forces within a large TP
rotating assembly and its large stationary housing assembly can be accomplished
by balance pistons. Some TPs have balancing pistons and some use the fluid
pressures on the back side of an impeller or turbine disk to make an effective
balance piston. Otherwise it would cause undesirable rubbing of certain parts
of the rotating TP assembly with parts of the stationary TP housing assembly.

378 TURBOPUMPS AND THEIR GAS SUPPLIES

Unbalanced large high-pressure loads cause a very small axial movement of the
rotating assembly (typically just 0.001 to 0.003 inch) and the balancing mech-
anism can be usually designed to prevent further axial movement and intense
rubbing of stationary and rotating parts. Pressure on the balance piston is con-
trolled by a slight axial movement, which changes the circular clearances at the
edge of the piston, and the rotating assembly is kept from moving more that
a couple of thousands of an inch. During the transient operations (start, stop,
change thrust) there typically are brief periods of unbalance caused by changes
in the pressure distributions on the internal surfaces normal to the shaft, but they
are of short duration. Nevertheless there can be some rubbing or axial contact of
a rotating and a stationery surface normal to the shaft during the brief periods
of unbalance. If the axial rubbing contact persists, there will usually be some
damage to the TP.

A phenomena characteristic of rotating machinery is the relative displacement
and very small movement of the center of the rotating shaft within the con-
finement of small clearances of a hxed sleeve bearing; this is known as whirl.
Reference 10-16 describes a condition where the relative small movement of the
shaft’s center is synchronous or the same as the rotation movement of the shaft
itself (revolutions per unit time).

The above equations relate to the steady-state operating condition. The tran-
sient and dynamic conditions are important and many have been analyzed using
iterative procedures and digital computers. Many have been tested, as shown
in Ref. 10-17. They are the start, stop, and thrust change. For example, the
transient starting condition includes the tank pressurization, the Tiling of pipes,
pumps, or manifolds with liquid propellants, the filling of turbines and their man-
ifolds with heated gas, ignition of gas generators, or thrust buildup. However, no
detailed discussion of these transient conditions will be given in this book. These
dynamic conditions can be complex, are related to the combustion behavior, and
are sometimes difficult to analyze. Each major rocket propulsion organization has
developed some method for analyzing these transients and they are often specific
to specific engines or specific organizations. A similar yet simpler and different
transient analysis of flows and pressure drops is usually performed for engines
with a pressurized gas feed system.

10.5. PUMPS

Classification and Description

The centrifugal pump is generally considered the most suitable for pumping
propellant in medium-sized and large rocket units. For the large flows and high
pressures involved, they are efficient as well as economical in terms of mass and
space requirement.

Figure 10-5 is a simplified schematic drawing of a centrifugal pump. Fluid
entering the impeller, which is essentially a wheel with spiral curved vanes rotat-
ing within a casing, is accelerated within the impeller channels and leaves the

10.5. PUMPS 379

Inlet flange

FIGURE 10-5. Simplified schematic half cross section of a typical centrifugal pump.

impeller periphery with a high velocity to enter the volute, or collector, and
thereafter the dijfuser, where conversion from kinetic energy (velocity) to poten-
tial energy (pressure) takes place. In some pumps the curved diffuser vanes are
upstream of the collector. The three-dimensional hydraulic design of impeller
vanes, diffuser vanes, and volute passages can be accomplished by computer
programs to give high efficiency and adequate strength. Internal leakage, or
circulation between the high-pressure (discharge) side and the low-pressure (suc-
tion) side of an impeller, is held to a minimum by maintaining close clearances
between the rotating and stationary parts at the seals or wear ring surfaces.
External leakage along the shaft is minimized or prevented by the use of a shaft
seal. Single-stage pumps (one impeller only) are stress limited in the pressure
rise they can impart to the liquid, and multiple-stage pumps are therefore needed
for high pump head,* such as with liquid hydrogen. References 10-2, 10-5,
10-8, 10-9, and 10-17 give information on different pumps. There is a free
passage of flow through the pump at all times, and no positive means for shutoff
are provided. The pump characteristics, that is, the pressure rise, flow, and effi-
ciency, are functions of the pump speed, the impeller, the vane shape, and the
casing configuration. Figure 10-6 shows a typical set of curves for centrifugal
pumps. The negative slope on the head versus flow curve indicates a stable pump
behavior. Reference 10-7 describes the development of a smaller turbopump and
the testing of a spiral high-speed first-stage impeller, called an inducer.

A shrouded impeller has a shroud or cover (in the shape of a surface of
revolution) on top of the vanes as shown in Figs. 10-1, 10-3, and 10-5. This
type usually has higher stresses and lower leakage around the impeller. In an

*Pump head means the difference between pump discharge and pump suction head. Its units are
meters or feet. The head is the height of a column of liquid with equivalent pressure at its bottom.
The conversion from pounds per square inch into feet of head is: (X) psi = 144 (,!(')/density (Ib/fO).
To convert Pascals (N/m^) of pressure into column height (m), divide by the density (kg/m^) and
go (9.806m/sec^).

jaModasjoq d>|ejg

I § I 8

I I I I I I I I

)33j ‘pean uoipns SAOjSod )3U psjjnbay

—I o

1 I I I I I I I I I I

88SSSS5gg2°

" tu 30 j 9 d duinj

FIGURE 10-6. Water test performance curves of the centrifugal pumps of the German V-2 rocket engine. The propellants are 75%
alcohol and liquid oxygen.

10.5. PUMPS 381

unshrouded impeller or turbine the vanes are not covered as seen in the turbine
vanes in Fig. 10-2.

Pump Parameters

This section outlines some of the important parameters and features that have
to be considered in the design of rocket propellant centrifugal pumps under
steady-flow conditions.

The required pump flow is established by the rocket engine design for a given
thrust, effective exhaust velocity, propellant densities, and mixture ratio. In addi-
tion to the flow required by the thrust chamber, the propellant consumption of
the gas generator, and in some designs also a bypass around the turbine and aux-
iliaries have to be considered in determining the pump flows. The required pump
discharge pressure is determined from the chamber pressure and the hydraulic
losses in valves, lines, cooling jacket, and injectors (see Eq. 10-7). To obtain
the rated flow at the rated pressure, an additional adjustable pressure drop for a
control valve or orifice is usually included which permits a calibration adjustment
or change in the required feed pressure. A regulation of the pump speed can also
change the required adjustable pressure drop. As described in Section 11.5, this
adjustment of head and flow is necessary to allow for hydraulic and performance
tolerances on pumps, valves, injectors, propellant density, and so on.

It is possible to predict the pump performance at various speeds if the perfor-
mance is known at any given speed. Because the fluid velocity in a given pump
is proportional to the pump speed N , the flow quantity or discharge Q is also
proportional to the speed and the head H is proportional to the square of the
speed. This gives the following relations:

Q(flow) ~ N (rpm or rad/sec)

77 (pump head) - (10-8)

P(pump power) ~

From these relations it is possible to derive a parameter called the specific
speed Ns- It is a dimensionless number derived from a dimensional analysis of
pump parameters as shown in Ref. 10-8.

Ns = Ns/&/(go A77,)3/4 (10-9)

Any set of consistent units will satisfy the equation: for example, N in radians
per second, Q in m^/sec, go as 9.8m/sec^, and 77 in meters. The subscript e refers
to the maximum efficiency condition. In U.S. pump practice it has become the
custom to delete go, express N in rpm, and Q in gallons per minute or ft^/sec.
Much of the existing U.S. pump data is in these units. This leads to a modified
form of Eq. 10-2, where Ns is not dimensionless, namely

A, = 21.2A7a/(A77,)3/4

( 10 - 10 )

382 TURBOPUMPS AND THEIR GAS SUPPLIES

TABLE 10-2. Pump Types

Impeller type

Radial Francis Mixed Flow Near Axial Axial

Basic shape
(half section)

Specific speed Ns
U.S. nomenclature
SI consistent units

Efficiency %

500-1000 1000-2000 2000-3000
0.2-0.3 0.4 0.6-0.8

50-80 60-90 70-92

3000-6000 Above 8000
1. 0-2.0 Above 2.5

76-88 75-82

Adapted from Ref. 10-8.

The factor 21.2 applies when N is in rpm, Q is in ft^/sec, and H is in feet. For
each range of specific speed, a certain shape and impeller geometry has proved
most efficient, as shown in Table 10-2. Because of the low density, hydrogen
can be pumped effectively by axial-flow devices.

The impeller tip speed in centrifugal pumps is limited by design and material
strength considerations to about 200 to 450 m/sec or roughly 655 to 1475 ft/sec.
With titanium (lower density than steel) and machined unshrouded impellers a
tip speed of over 2150 ft/sec is now possible and used on the pump shown in
Fig. 10-2. For cast impellers this limiting value is lower than for machined
impellers. This maximum impeller tip speed determines the maximum head that
can be obtained from a single stage. The impeller vane tip speed u is the product
of the shaft speed, expressed in radians per second, and the impeller radius and
is related to the pump head A.H by

u = ir^lgQ AH (10-11)

where the velocity factor tj/ has values between 0.90 and 1.10 for different
designs. For many pumps, = 1.0.

The volume flow Q defines the impeller inlet and outlet areas according to the
equation of continuity. The diameters obtained from this equation should be in
the proportion indicated by the diagrams for a given specific speed in Table 10-2.
The continuity equation for an incompressible liquid is

Q=Aivi=A 2 V 2 (10-12)

where the subscripts refer to the impeller inlet and outlet sections, all areas A
being measured normal to their respective flow velocity v. The inlet velocity m
ranges usually between 2 and 6 m/sec or 6.5 to 20 ft/sec and the outlet velocity

10.5. PUMPS 383

V 2 between 3 and 15 m/sec or 10 to 47 ft/sec. For a compressible liquid, such as
liquid hydrogen, the density p will change with pressure. The continuity equation
then is in terms of mass flow in

m = Aivipi = A 2 V 2 P 2 (10-13)

The head developed by the pump will then also depend on the change in density.

The pump performance is limited by cavitation, a phenomenon that occurs
when the static pressure at any point in a fluid flow passage becomes less than
the fluid’s vapor pressure. Cavitation is discussed in Refs. 10-1 to 10-5, 10-8,
10-9, and 10-17. The formation of vapor bubbles in the low-pressure regions
causes cavitation. These bubbles collapse when they reach a region of higher
pressure, that is, when the static pressure in the fluid is above the vapor pressure.
In centrifugal pumps cavitation is most likely to occur behind the leading edge of
the pump impeller vane at the inlet because this is the point at which the lowest
absolute pressure is encountered. The excessive formation of vapor causes the
pump discharge mass flow to diminish and fluctuate and can reduce the thrust
and make the combustion erratic and dangerous.

When the bubbles travel along the pump impeller surface from the low-
pressure region (where they are formed) to the downstream higher-pressure
region, the bubbles collapse. The sudden collapses create local high-pressure
pulses that have caused excessive stresses in the metal at the impeller surface.
In most rocket applications this cavitation erosion is not as serious as in water
or chemical pumps because the cumulative duration is relatively short and the
erosion of metal on the impeller is not usually extensive. However, it has been
a concern with some test facility transfer pumps.

The required suction head {Hs)r is the limit value of the head at the pump inlet
(above the local vapor pressure); below this value cavitation in the impeller may
occur. It is a function of the pump and impeller design and its value increases
with flow as can be seen in Fig. 10-6. To avoid cavitation the suction head
above the vapor pressure required by the pump {Hs)r must always be less than
the available or net positive suction head furnished by the line up to the pump
(Hs)a, that is, {Hs)r < {Hs)a- The required suction head above vapor pressure
can be determined from the suction speciflc speed S :

S =2l.2N^e/(H,)T (10-14)

The suction speciflc speed S depends on the quality of design and the speciflc
speed Ns, as shown in Table 10-2. The suction speciflc speed S has a value
between 5000 and 60,000 when using ft-lbf units. For pumps with poor suc-
tion characteristics it has values near 5000, for the best pump designs without
cavitation it has values near 10,000 and 25,000, and for pumps with limited
and controllable local cavitation it has values above 40,000. In Eq. 10-14 the
required suction head {Hs)r is usually defined as the critical suction head at
which the developed pump discharge head has been diminished arbitrarily by

384 TURBOPUMPS AND THEIR GAS SUPPLIES

2% in a pump test with increasing throttling in the suction side. Turbopump
development has, over the last several decades, led to impeller designs which
can operate successfully with considerably more cavitation than the arbitrary and
commonly accepted 2% head loss limit. Inducers are now designed to run stably
with extensive vapor bubbles near the leading edge of their vanes, but these bub-
bles collapse at the trailing end of these vanes. Inducers now can have S values
above 80,000. A discussion of one method for the design of impeller blades can
be found in Ref. 10-8.

The head that is available at the pump suction flange is called the net positive
suction head or available suction head above vapor pressure {Hs)a- It is an
absolute head value determined from the tank pressure (the absolute gas pressure
in the tank above the liquid level), the elevation of the propellant level above
the pump inlet, diminished by the friction losses in the line between tank and
pump, and the vapor pressure of the fluid. When the flying vehicle is undergoing
accelerations, the head due to elevation must be corrected accordingly. These
various heads are defined in Fig. 10-7. The net positive suction head (Hs)a is
often abbreviated as NPSH and is the maximum head available for suppressing
cavitation at the inlet to the pumps;

{bls)A — ^tank “f ^elevation ^friction ^vapor (10—15)

Tank pressure
gauge (SJ

Fluid

level

Atmospheric
pressure

^ Absolute tank_
pressure, Huuj,

Gauge tank
gas pressure

Typical |i/=^
valve “>=

Pump inlet

Friction loss Hr

Absolute static
head at pump
inlet with
zero flow

Elevation

head

^ Atmospheric
^ pres^sure

Friction loss

Vapor pressure
of fluid,

Available gauge
pressure at
pump inlet
with flow

Absolute available
dynamic and
static head at
pump inlet
with flow

Net positive suction
head (NPSH) or
maximum head
available for
surpressing cavitation
In the pump
(Hs)x

flange

Pressure reference line

Pump

FIGURE 10-7. Definition of pump suction head.

10.5. PUMPS 385

To avoid pump cavitation, {Hs)a has to be higher than (Hs)r. If additional head
is required by the pump, the propellant may have to be pressurized by external
means, such as by the addition of another pump in series (called a booster pump)
or by gas pressurization of the propellant tanks. This latter method requires thicker
tank walls and, therefore, heavier tanks, and a bigger gas-pressurizing system.
For example, the oxygen tank of the German V-2 was pressurized to 2.3 atm,
partly to avoid pump cavitation. For a given value of {Hs)a, propellants with
high vapor pressure require correspondingly higher tank pressures and heavier
inert tank masses. For a given available suction head {Hs)a, a pump with a low
required suction pressure usually permits designs with high shaft speeds, small
diameter, and low pump inert mass. A small value of {Hs)r is desirable because it
may permit a reduction of the requirements for tank pressurization and, therefore,
a lower inert tank mass. The value of (Hs)r will be small if the impeller and
fluid passages are well designed and if the shaft speed N is low. A very low
shaft speed, however, requires a large-diameter pump, which will be excessively
heavy. The trend in selecting centrifugal pumps for rocket application has been
to select the highest shaft speed that gives a pump with a low value of (Hs)r,
does not require excessive tank pressurization or other design complications, and
thereby permits relatively lightweight pump design. This places a premium on
pumps with good suction characteristics.

There have been some low-thrust, low-flow, experimental engines that have
used positive displacement pumps, such as diaphragm pumps, piston pumps, or
rotary displacement pumps (gear and vane pumps). For low values of specific
speed Ns these pumps have much better efficiencies, but their discharge pressures
fluctuate with each stroke and they are noisy.

One method to provide a lightweight turbopump with low vehicle tank pressure
is to use an inducer, which is a special pump impeller usually on the same shaft
and rotating at the same speed as the main impeller. It has a low head rise
and therefore a relatively high specific speed. Inducer impellers are immediately
upstream of the main impeller. They are basically axial-flow pumps with a spiral
impeller, and many will operate under slightly cavitating conditions. The inducer
stage’s head rise (typically, 2 to 10% of the total pump head) has to be just large
enough to suppress cavitation in the main pump impeller; this allows a smaller,
lighter, higher-speed main pump. Figures 10-2 and 10-8 and Reference 10-18
show inducers. Reference 10-19 describes the testing of a pump with an inducer.
Most TPs today have inducers.

Influence of Propellants

For the same power and mass flow, the pump head is inversely proportional to
the propellant density. Since pumps are basically constant-volume flow machines,
the propellant with the highest density requires less head, less power, and thus
allows a smaller pump assembly.

Because many of the propellants are dangerous to handle, special provi-
sions have to be made to prevent any leakage through the shaft seals. With

386

TURBOPUMPS AND THEIR GAS SUPPLIES

FIGURE 10-8. Fuel pump inducer impeller of the Space Shuttle main engine low-
pressure fuel turbopump. It has a diameter about 10 in., a nominal hydrogen flow of
148.6 Ibm/sec, a suction pressure of 30 psi, a discharge pressure of 280 psi at 15,765
rpm, an efficiency of 77%, and a suction specific speed of 39,000 when tested with water.
(Courtesy of Pratt & Whitney Rocketdyne.)

spontaneously ignitable propellants the leakages can lead to fires in the pump
compartment and may cause explosions. Multiple seals are often used with a
drainage provision that safely removes or disposes of any propellants that flow
past the first seal. Inert-gas purges of seals have also been used to remove haz-
ardous propellant vapors. The sealing of corrosive propellants puts very severe
requirements on the sealing materials and design. With cryogenic propellants
the pump bearings are usually lubricated by the propellant, since lubricating oil
would freeze at the low pump hardware temperature.

Centrifugal pumps should operate at the highest possible pump efficiency. This
efficiency increases with the volume flow rate and reaches a maximum value of
about 90% for very large flows (above 0.05 m^/sec) and specific speeds above
about 2500 (see Refs. 10-1 and 10-8). Most propellant pump efficiencies are
between 30 and 70%. The pump efficiency is reduced by surface roughness
of casing and impellers, the power consumed by seals, bearings, and stuffing
boxes, and by excessive wear ring leakage and poor hydraulic design. The pump
efficiency rjp is defined as the fluid power divided by the pump shaft power Pp -.

r]p=pQAH/Pp (10-16)

A correction factor of 550 ft-lbf/hp has to be added if Pp is given in horsepower,
the head H in feet, and the volume flow Q in ft^ /sec. When using propellants,
the pump power has to be multiplied by the density ratio if the required power
for water tests is to be determined.

10 . 6 . TURBINES 387

Example 10-1. Determine the shaft speed and the overall impeller dimensions for a
liquid oxygen pump which delivers 500 Ibm/sec of propellant at a discharge pressure of
1000 psia and a suction pressure of 14.7 psia. The oxygen tank is pressurized to 35 psia.
Neglect the friction in the suction pipe and the suction head changes due to acceleration
and propellant consumption. The initial tank level is 15 ft above the pump suction inlet.

SOLUTION. The density of liquid oxygen is 71.2 Ibm/ft^ at its boiling point (from
Chapter 7). The volume flow will be 500/71.2 = 7.022 ft^/sec. The vapor pressure of the
oxygen is 1 atm = 14.7 psi = 29.8 ft. The suction head is 35 x 144/71.2 = 70.8 ft. From
Eq. 10-12 the available suction head is 70.8 + 29.8 = 100.6 ft. The available suction head
above vapor pressure is {Hs)a = 70.8 + 15.0 — 0 — 29.8 = 56.0 ft. The discharge head
is 1000 X 144/71.2 = 2022 ft. The head delivered by the pump is then 2022 — 100.6 =
1921 ft.

The required suction head will be taken as 80% of the available suction head in order
to provide a margin of safety for cavitation (Hs)r = 0.80 x 100.6 = 80.48 ft. Assume a
suction specific speed of 15,000, a reasonable value if no test data are available. From
Eq. 10-14 solve for the shaft speed N\

S = = 21.2AV7.022/80.48°''^ = 15,000

Solve for N = 7174 rpm or 751.3 rad/sec.

The specific speed, from Eq. 10-9, is

A, = 2\.2Ny[QIH^I‘' = 21.2 x 7174V7.022/192l“-’^ = 1388

According to Table 10-2, the impeller shape for this value of A, will be a Francis type.
The impeller discharge diameter D 2 can be evaluated from the tip speed by Eq. 10-11:

u = fy/2go AH = I.OV 2 X 32.2 x 1921 = 352 ft/sec

D 2 = 353 X 2/664.7 = 1.062 ft = 12.75 in.

The impeller inlet diameter Dj can be found from Eq. 10-5 by assuming a typical inlet
velocity of 15 ft/sec and a shaft cross section 5.10 in.^ (2.549 in. diameter).

A = Q/vi = 7.022/15 = 0.468 ft^ = 67.41 in.^

A = ijrDf = 67.41 + 5.10 = 72.51 in.^

Di = 9.61 in. (internal flow passage diameter)

This is enough data to draw a preliminary sketch of the impeller.

10.6. TURBINES

The turbine must provide adequate shaft power for driving the propellant pumps
(and sometimes also auxiliaries) at the desired shaft speed and torque. The turbine
derives its energy from the expansion of a gaseous working fluid through fixed

388 TURBOPUMPS AND THEIR GAS SUPPLIES

nozzles and rotating blades. The blades are mounted on disks to the shaft. The gas
is expanded to a high, nearly tangential, velocity and through inclined nozzles and
then flows through specially shaped blades, where the gas energy is converted
into tangential forces on each blade. These forces cause the turbine wheel to
rotate (see Refs. 10-1 to 10-5, 10-20 and 10-21).

Classification and Description

The majority of turbines have blades at the periphery of a turbine disk and the
gas flow is axial, similarly in concept to the axial-flow pattern shown for pumps
in Table 10-2 and the single-stage turbine of Fig. 10-1. However, there are a
few turbines with radial flow (particularly at high shaft speeds), such as the one
shown in Fig. 10-2. Ideally there are two types of axial-flow turbines of interest
to rocket pump drives: impulse turbines and reaction turbines, as sketched in
Fig. 10-9. In an impulse turbine the enthalpy of the working fluid is converted
into kinetic energy within the first set of stationary turbine nozzles and not in
the rotating blade elements. High-velocity gases are delivered (at a small angle
to a tangential direction) to the rotating blades, and blade rotation takes place as
a result of the impulse imparted by the momentum of the fluid stream of high
kinetic energy to the rotating blades which are mounted on the turbine disk. The
velocity-staged impulse turbine has a stationary set of blades which changes the
flow direction after the gas leaves the first set of rototating blades and directs
the gas to enter a second set of rotating blades in which the working fluid gives
up further energy to the turbine wheel. In a pressure-staged impulse turbine, the
expansion of the gas takes place in all the stationary rows of blades. In a reaction
turbine the expansion of the gas is roughly evenly split between the rotating and
stationary blade elements. The high-pressure drop available for the expansion of
the turbine working fluid in a gas generator cycles favors simple, lightweight
one- or two-stage impulse turbines for high thrust engines. Many rocket turbines
are neither pure impulse nor reaction turbines, but often are fairly close to an
impulse turbine with a small reaction in the rotating vanes.

With some gas generator engine cycles the turbine exhaust gases pass through
a De Laval Supersonic nozzle at the exit of the exhaust pipe (see Fig. 1-4). The
high turbine outlet pressure gives critical flow conditions at the venturi throat
(particularly at high altitudes) and thereby assures a constant turbine outlet pres-
sure and a constant turbine power which will not vary with altitude. Furthermore,
it provides a small additional thrust to the engine.

Turbine Performance and Design Considerations

The power supplied by the turbine is given by a combined version of Eqs. 3-1
and 3-7:

P, = t]Tmt/Sh (10-17)

P, = rnmrCpTdl - (10-18)

10 . 6 . TURBINES 389

impulse turbine

FIGURE 10-9. Top view diagram, pressure and velocity profiles, and efficiency curves
for impulse and reaction type turbines. The velocity ratio is the pitch line velocity of the
rotating blades u divided by the theoretical gas spouting velocity co derived from the
enthalpy drop. (Adapted with permission from Refs. 10-3 and 10-19.)

The power delivered by the turbine Pt is proportional to the turbine efficiency
rit, the mass flow through the turbine tii,, and the available enthalpy drop per
unit of flow A/t. The units in this equation have to be consistent (1 Btu =
778 ft-lbf = 1055 J). This enthalpy is a function of the specific heat Cp, the
nozzle inlet temperature Ti, the pressure ratio across the turbine, and the ratio of
the specific heats k of the turbine gases. For gas generator cycles the pressure
drop between the turbine inlet and outlet is relatively high, but the turbine flow
is small (typically 2 to 5% of full propellant flow). For staged combustion cycles
this pressure drop is very much lower, but the turbine flow is much larger.

For very large liquid propellant engines with high chamber pressure the turbine
power can reach over 250,000 hp, and for small engines this could be perhaps
around 35 kW or 50 hp.

390 TURBOPUMPS AND THEIR GAS SUPPLIES

According to Eq. 6-12, the power delivered by the turbine Pt has to be
equal to the power required by the propellant pumps, the auxiliaries mounted on
the turbopump (such as hydraulic pumps, electric generators, tachometers, etc.),
and power losses in hydraulic friction, bearings, gears, seals, and wear rings.
Usually these losses are small and can often be neglected. The effect of the
turbine gas flow on the specific impulse of the rocket engine system is discussed
in Section 6.6. For gas generator engine cycles, the rocket designer is interested
in obtaining a high turbine efficiency and a high turbine inlet temperature T\ in
order to reduce the flow of turbine working fluid, and for gas generator cycles
also to raise the overall effective specific impulse, and, therefore, reduce the
propellant mass required for driving the turbine. Three-dimensional computer
analyses of the gas flow behavior and turbine blade geometry have resulted in
efficient blade designs.

Better turbine blade materials (such as single crystals which have been uni-
directionally solidified) and specialty alloys can allow turbine inlet temperatures
between 1400 K (or about 2050°F) and perhaps 1600 K (or 2420°F); these higher
temperatures or higher gas enthalpies reduce the required turbine flow. Reliabil-
ity, gas temperature variations or nonuniformity, and cost considerations have
kept actual turbine inlet temperatures at conservative values, such as 1150 to
1250°F or about 900 to 950 K, using special steel alloys as the blade and disk
materials. The efficiency of turbines for rocket turbopumps is shown in Fig. 10-9.
Maximum blade speeds with good design and strong high-temperature materials
are typically 400 to 700 m/sec or about 1300 to 2300 ft/sec. Fligher blade speeds
generally allow an improvement in efficiency. For the efficiency to be high the
turbine blade and nozzle profiles have to have smooth surfaces. Small clearances
at the turbine blade tips are also needed to minimize leakage around the blades.

Different organizations define turbine efficiency in different ways. One way
to define the turbine efficiency ip is the turbine power output LN divided by the
turbine input from an ideal uniform heated gas flowing through an ideal turbine.
The gas flow can be considered the ideal isentropic enthalpy drop. The actual
pressure drop and the actual Ah or usually higher than the ideal:

T], =L,N,/rn Ah (10-19)

Flere Ah is the enthalpy drop with isentropic gas expansion across the turbine
nozzles and the turbine buckets or blades (moving and stationary) per unit gas
mass flow. It is typically based on a uniform gas flow across a flow section,
uniform gas properties across a flow section, perfect gases, no leakage or alter-
natively a nominal leakage around the blades, and perfect blade contours. Even a
small clearance at the tip of turbine blades causes substantial losses, particularly
at small turbine diameter and small blade heights. Some organizations include a
few unavoidable basic losses in defining the ideal power.

The advantage of increased turbine efficiency (less gas generator propellant
requirement) can be realized only if the turbopump design has efficient turbine
blade contours and allows high blade speeds. This speed can be achieved in

10.7. APPROACH TO TURBOPUMP PRELIMINARY DESIGN 391

rockets of medium and low thrust by gearing the turbine to the pump shaft or
by using pumps that permit high shaft speeds; in rockets of very high thrust the
pumps have diameters and shaft speeds close to those of the turbines and can
be mounted on the same shaft as the turbine. The power input to the turbine can
be regulated by controlling the gas flow and gas temperatures to the turbine inlet
through valves or orifices.

Usually, the preliminary analysis for the pump is done first. Avoiding exces-
sive cavitation sets a key pump parameter, namely the maximum pump shaft
speed. This is the highest possible shaft speed, which in turn allows the lightest
turbopump mass, without excessive cavitation in the pump. If excessive cavita-
tion occurs at the leading edge of the main impeller, then the flow will become
unsteady and variable, leading to lower thrust and possible combustion instabil-
ity. The amount of pressure in the vehicle (gas pressure in propellant tank plus
the static elevation pressure) that can be made available to the engine (at the
pump inlet) for suppressing cavitation has to be larger than the impeller vanes’
own pressure limit to cavitate. This allows us then to determine the shaft speed,
which in turn can establish the approximate pump efficiencies, impeller tip speed
(usually also limited by the material strength of the impeller), number of pump
stages, key dimensions of the impeller.

10.7. APPROACH TO TURBOPUMP PRELIMINARY DESIGN

With all major rocket engine components the principal criteria (high performance
or efficiency, minimum mass, high reliability, and low cost) have to be weighted
and prioritized for each vehicle mission. For example, high efficiency and low
mass usually mean low design margins, and thus lower reliability. A higher
shaft speed will allow a lower mass turbopump, but it cavitates more readily
and requires a higher tank pressure and heavier vehicle tanks (which usually
outweighs the mass savings in the turbopump).

The engine requirements give the initial basic design goals for the preliminary
design of the turbopump, namely propellant flow, the pump outlet or discharge
pressure (which has to be equal to the chamber pressure plus the pressure drops in
the piping, valves, cooling jacket, and injector), the best engine cycle (as shown
in Fig. 6-9), the start delay, and the need for restart or throttling, if any. Also, the
propellant properties (density, vapor pressure, viscosity, or boiling point) must
be known. Some of the design criteria are explained in Refs. 10-2 and 10-3,
and basic texts on turbines and pumps are listed as Refs. 10-4 to 10-8.

There are several design variations or geometrical arrangements for transmit-
ting turbine power to one or more propellant pumps; some are shown schemati-
cally in Fig. 10-4 and an initial selection has to be made. If there is a mismatch
between the optimum pump speed and the optimum turbine speed (which is usu-
ally higher), it may save inert mass and turbine drive gas mass to interpose a gear
reduction between their shafts. See Fig. 6-11 and Figs. 10-4d and 10-4T For
the last two decades designers have preferred to use direct drive, which avoids

392 TURBOPUMPS AND THEIR GAS SUPPLIES

the complication of a gear case but at a penalty in efficiency and the amount of
turbine drive propellant gas required. See Figs. 6-12, 10-1, 10-2, and 10-4e.

The key turbine parameter can be estimated because the power output of the
turbine essentially has to equal the power demand of the pumps. If the pump is
driven directly, that is, without a gear case, then the pump speed and the tur-
bine speed are equal. From the properties of the turbine drive gas (temperature,
specific heat, etc.), the strength limits of the turbine materials, and the likely
pressure drop, it is possible to determine the basic dimensions of the blades
[pitch line velocity, turbine nozzle outlet velocity, number of rows (stages) of
blades, turbine type, or turbine efficiency]. The particular arrangement or geom-
etry of the major turbopump components is related to their selection process.
Most propellant pumps have a single-stage main impeller. For liquid hydrogen
with its low density, a two- or three-stage pump is normally needed. Usually
some design limit is reached which requires one or more iterations, each with a
new changed approach or parameter. The arrangement of the major turbopump
components (Fig. 10-4) is also influenced by the position of the bearings on the
shaft. For example, we do not want to place a bearing in front of an impeller inlet
because it will cause turbulence, distort the flow distribution, raise the suction
pressure requirement, and make cavitation more likely to occur. Also, bearings
positioned close to a turbine will experience high temperatures, which influences
the lubrication by propellant and may demand more cooling of the bearings.

The use of booster pumps allows lower tank pressure, and thus lower inert
vehicle mass, and provides adequate suction pressures to the main pump inlet.
See Ref. 10-12. Booster pumps are used in the Space Shuttle main engine and
the Russian RD-170, as seen in Figs. 6-12 and 11-2. Some booster pumps
have been driven by a liquid booster turbine using a small flow of high-pressure
liquid propellant that has been tapped off the discharge side of the main pump.
The discharged turbine liquid then mixes with the main propellant flow at the
discharge of the booster pump.

Mass is at a premium in all flying vehicle installations, and the feed system
is selected to have a minimum combined mass of tubines, pumps, gas generator,
valves, tanks, and gas generator propellants. Some of the considerations in the
design of turbopumps are the thermal stresses, warpage due to thermal expansion
or contraction, axial loads, adequate clearances to prevent rubbing yet minimize
leakage, alignment of bearings, provisions for dynamic balancing of rotating
parts, mounting on an elastic vehicle frame without inducing external forces, and
avoiding undue pressure loads in the liquid and gas pipes.

Vibrations of turbopumps have caused problems during development. The
analyses of the various vibrations of components (shaft, turbine blades, liquid
oscillations, gas flow oscillations, or bearing vibrations) are not given here. At the
critical speed the natural structural resonance frequency of the rotating assembly
(shaft, impellers, turbine disk, etc.) coincides with the rotation operating speed,
a condition that must be minimized. A slight unbalance can then be amplified to
cause significant shaft deflections, bearing failure, and other damage. The oper-
ating speed therefore is usually lower, but occasionally higher than the critical

10.7. APPROACH TO TURBOPUMP PRELIMINARY DESIGN 393

speed. A large diameter stiff shaft, rigid bearings, and stiff bearing supports will
increase this critical speed, and damping (such as the liquid lubricant him in
the bearing) will reduce the vibration amplitude. The solving of various internal
vibration problems, such as whirl in bearings and blade vibrations, is reported
in Ref. 10-16 and the dynamics of the propellant flow in pumps in Ref. 10-15.
Vibrations of adjacent components can be excited by a TP and the natural fre-
quency of these components can be analyzed. If the operating frequency or
the shaft speed of the rotating assembly coincides with the natural frequency,
another engine part, such as injector domes or certain piping assemblies, then an
undesirable resonance with excessive stresses can occur. As a result the natural
frequencies of affected component or of the TP can be changed by design. For
example, whenever the tip of a pump blade goes by the tongue of the pump
volute, a pressure wave is generated in the liquid. The frequency is the product
of the number of pump vanes and the speed of the vane tip. Vibrations external
to the TP, such as those generated by a thrust chamber (see Chapter 9) should not
influence the operation of a TP. This interaction, if it occurs, can be determined
by hot fire tests of a complete engine with appropriate instruments measuring TP
behavior.

The bearings in most existing TPs are high precision, special metal alloy ball
bearings (Ref. 10-13). A few are roller bearings, which have a higher radial
load capacity. Some ball bearings can take both radial and axial loads. Early
ball and roller bearings were limited in the loads and speeds at which they
could operate reliably. New bearings were developed by the bearing industry. In
some TP designs the limits of the bearing loads and speeds determined the shaft
speed and thus the minimum size of the turbopump, rather than the cavitation
limit of the pump. Various types of bearings, and many types of seals have
been explored, tested in laboratory fixtures or experimental TPs. This included
hydrostatic (precision sleeve type) bearings, foil bearings, and magnetic bearings
(Ref. 10-13). As far as the authors know, none of these have as yet found their
way into a production TP. The variety of static and dynamic seals is relatively
large, and the selection is strongly influenced by the preferences of the design
organization (Ref. 10-14). One type of bearing oscillation problem is discussed
an Ref 10-16.

If the turbopump is part of a reusable rocket engine, it becomes more complex.
For example, it can include provision to allow for inspection and automatic
condition evaluation after each mission or flight. This can include an inspection
of bearings through access holes for horoscope instruments, checking for cracks
in highly stressed parts (turbine blade roots or hot-gas high-pressure manifolds),
or the measurement of shaft torques (to detect possible binding or warpage).

The number of different materials of construction of TP appears to have
increased. For example, for high-speed, high-load, ball bearings a new ball
material (silicon nitride) has been successfully introduced. Relatively common
materials, such as stainless steels, have in part been replaced by superalloys,
such as Inconel. Powder metallurgy has found its way into impellers and turbine
parts. Although the strength is not really better than a forged or cast material.

394 TURBOPUMPS AND THEIR GAS SUPPLIES

they have a smooth surface (low friction), uniform physical properties, and can
be fabricated in complex shapes (Ref. 10-22).

There is no warm-up time available in rocket turbines. The sudden admission
of hot gas at full flow causes severe thermal shock and thermal distortion and
increases the chances for rubbing between moving metal parts. The most severe
stresses of a turbine blade often are thermal stresses; they come during the engine
start when the leading edge is very hot but other parts of the blade are still cold.

For low-thrust engines the shaft speeds can become very high, such as over
100,000 rpm. Also, the turbine blade height becomes very short and friction
losses can become prohibitive. In order to obtain a reasonable blade height we
go to partial admission turbine designs. Here a portion of the turbine nozzles are
effectively plugged or eliminated.

10.8. GAS GENERATORS AND PREBURNERS

The purpose of a gas generator or a preburner is to create the “warm” gas (usually
between 600 and 2000°F or 315 to 1200°C) to drive the turbine of a TP. See
Refs. 10-1 to 10-5. The selected gas temperature depends on nearly uniform
gas composition, the engine design, the turbine blade materials, using uncooled
hardware, and the acceptable risk of failure. In order to achieve these warm gas
temperatures with a bipropellant, most operate at a fuel-rich mixture ratio, but
some of the prebumers in the Soviet Union (today Russia) operate at an oxidizer-
rich mixture usually with engines using staged combustion engine cycles. Each
of these combustion devices consists of a combustion chamber, an injector, and
a pipe or duct leading to the turbine. They all have separate dedicated propellant
control valves, sometimes with some calibrated orifices, for control of the flows.
They resemble a thrust chamber, but they do not have a single bow-tie-shaped
supersonic nozzle. Instead the warm gas flows subsonically from the gas generator
or the prebumer through pipes or manifolds into the nozzles of the turbine (here
they reach high velocities) and then they flow supersonically through the row(s)
of turbine blades, which extract the energy for driving the propellant pumps.
Requirements for gas generators or preburners include the delivery of warm
gas at the intended mass flows, pressure, and design temperature, an essentially
uniform gas temperature across the flow path to the turbine and no high gas
temperature spikes. Combustion instability problems are extremely rare.

Gas generators are used exclusively with liquid propellant rocket engines,
which operate on a gas generator engine cycle, and preburners are used exclu-
sively with rocket engines, which operate on a staged combustion cycle and they
usually operate at a higher pressure. Table 10-3 explains some of the differences
between these two and Table 6-6 and Fig. 6-9 describe the engine cycles.

Gas generators are shown as a component of an engine in Fig. 1-4 and
Ref. 10-4 and prebumers in Figs. 6-1, 6-12, and 11-2. Propellants supplied
to the gas generator or the prebumer usually are tapped off from discharges of
the engines main pumps. When starting an engine, the turbomachinery needs to

PROBLEMS 395

TABLE 10-3. Comparison of Key Characteristics of Gas Generators and Prebumers

Parameter

Gas Generator

Preburner

Engine cycle

Gas generator cycle

Staged combustion cycle

Chamber pressure

Usually equal or lower than

30-60% higher than its

thrust chamber pressure

Mass flow as % of total
propellant flow

1-7

40-85

Cooling

Usually uncooled

Usually uncooled, but may
be partially cooled

Inert mass

Relatively light

Heavy

Size

Relatively small

Can be large

be brought up to speed before propellant can be supplied at pressure. This gas
generator start has also been done with a solid propellant starting cartridge (runs
only a few seconds), an auxiliary set of small propellant tanks pressurized by
cold gas (also runs only for short duration), or by letting the engine “bootstrap”
itself into a start using the modest tank pressure augmented by the liquid column
head existing in the vehicle tanks and feed system pipe lines — usually called
“tank head” start (requires more time to start). A discussion of engine starts
and tank pressurization can be found in Section 6.5 and thrust chamber starts in
Section 8.6.

In the past monopropellant gas generators were common and both 80 or 90%
hydrogen peroxide or pure hydrazine provided warm gas usually through catalytic
decomposition in a bed of solid catalysts. This type has a simpler gas generator
system (only one tank and one set of valves, instead of two), has no mixture ratio
adjustment, and a predictable, fully reproducible uniform warm gas temperature
without potential temperature spikes. The key disadvantages were the complica-
tions of providing a third propellant, the potential propellant hazards, the lower
performance, or the higher mass of propellant for making enough warm gas for
providing the required power.

PROBLEMS

1. A rocket engine with two TPs delivers the fuel, namely UDMH, at a pump discharge
pressure of 555 psia, a flow of 10.2 Ih/sec, at 3860 rpm, and a fuel temperature of
68°F. Determine the following:

(a) The fuel pump power for these nominal conditions.

(b) When the fuel flow is reduced to 70% of nominal, what will be the approximate
power level, discharge pressure, and shaft speed? Assume that the oxidizer pump is
also reduced by 70% and so is the gas flow to the turbines, but the gas temperature
is unchanged.

396 TURBOPUMPS AND THEIR GAS SUPPLIES

(c) If the allowed temperature variation of the propellants is at — 40°F and on another
day +120°F, how will this affect the power level, shaft speed, and the discharge
pressure of the fuel pump?

2. What are the specihc speeds of the four SSME pumps? (See the data given in
Table 10-1.)

3 . Compute the turbine power output for a gas consisting of 64% by weight of H 2 O and
36% by weight of O 2 , if the turbine inlet is at 30 atm and 658 K with the outlet at 1.4
atm and with 1.23 kg flowing each second. The turbine efficiency is 37%.

4 . Compare the pump discharge gauge pressures and the required pump powers for five
different pumps using water, gasoline, alcohol, liquid oxygen, and diluted nitric acid.
The respective specific gravities are 1.00, 0.720, 0.810, 1.14, and 1.37. Each pump
delivers 100 gal/min, a head of 1000 ft, and arbitrarily has a pump efficiency of 84%.
Answers: 433, 312, 350, 494, and 594 psi; 30.0, 21.6, 24.3, 34.2, and 41.1 hp.

5. The following data are given on a liquid propellant rocket engine:

Thrust

Thrust chamber specific impulse

Fuel

Oxidizer

Thrust chamber mixture ratio
Turbine efficiency
Required pump power
Power to auxiliaries mounted on turbopump
gear case

Gas generator mixture ratio
Turbine exhaust pressure
Turbine exhaust nozzle area ratio
Enthalpy available for conversion in turbine
per unit of gas

Specific heat ratio of turbine exhaust gas

40,200 Ibf
210.2 sec

Gasoline (specific gravity 0.74)
Red fuming nitric acid (sp. gr.

1.57)

3.25

58%

580 hp
50 hp

0.39
37 psia
1.4

180 Btu/lb
1.3

6. Determine the engine system mixture ratio and the system specific impulse.
Answers: 3.07 and 208 sec.

SYMBOLS

a gear ratio

A area, m^(ft^)

Cp specific heat at constant pressure, J/kg K (Btu/lb°R)

D diameter, m (ft)

go sea-level acceleration of gravity, 9.806 m/sec^ (32.17 ft/sec^ )

H head, m (ft)

(Hs)a available pump suction head above vapor pressure, often called

net positive suction head, m (ft)

{Hs)r required pump suction head above vapor pressure, m (ft)

REFERENCES 397

L torque, Nm (ft-lbf)

m mass flow rate, kg/sec (Ibm/sec)

N shaft speed, rpm (rad/sec)

Ns specific speed of pump

P pressure, N/m^ (Ibf/in.^ )

P power, W (hp)

P), power of auxiliaries, bearings, rubbing seals, friction

Q volume flow rate, m^/sec (ft^/sec)

S suction specific speed of pump

T absolute temperature, K (°R)

u impeller tip speed or mean blade speed, m/sec (ft/sec)

V liquid fiow velocity, m/sec (ft/sec)

Greek Letters

A finite differential

ri efficiency

p density, kg/m^(lb/ft^)

velocity correction factor

Subscripts

e maximum efficiency condition

/ fuel

o oxidizer

p pump

t turbine

0 initial condition

1 inlet

2 outlet

REFERENCES

10-1. G. P. Sutton, “Turbopumps, a Historical Perspective,” AIAA Paper 2006-7531,
July 2006.

10-2. D. K. Huzel and D. H. Huang, “Design of Turbopump Feed Systems,” Chapter 6
in Design of Liquid Propellant Rocket Engines, rev. ed., Vol. 147, Progress in
Astronautics and Aeronautics , AIAA, Reston, VA, 1992.

10-3. M. L. Strangeland, “Turbopumps for Liquid Rocket Engines,” Threshold, an Engi-
neering Journal for Power Technology , No. 3, Rocketdyne Propulsion and Power,
Summer 1988, pp 34-42.

10-4. G. P. Sutton, “Turbopumps,” Chapter 4.4 and “Gas Generators, Prebumers and
Tank Pressurization,” Chapter 4.5 in Elistory of Liquid Propellant Rocket Engines ,
AIAA, Reston, VA, 2006.

398

TURBOPUMPS AND THEIR GAS SUPPLIES

10-5. Turbopump Systems for Liquid Rocket Engines, NASA Space Vehicle Design
Monograph, NASA SP-8107, August 1974.

10-6. Personal communications with personnel from Pratt & Whitney Rocketdyne,
Northrop Grumman, and The Aerospace Corporation, 2006 to 2008.

10-7. A. Minick and S. Peery, “Design and Development of an Advanced Liquid Hydro-
gen Turbopump,” AlAA paper 98-3681, July 1998, and G. Crease, R. Lyda,
J. Park, and A. Minick, “Design and Test Results of an Advanced Liquid Hydro-
gen Pump,” AlAA paper 99-2190, 1999.

10-8. 1. Kassarik, W. C. Krutzsch, W. H. Frazer, and J. P. Messina (Eds.), Pump
Handbook, McGraw-Hill Book Company, New Tork, 1976 (water hammer and
pumps).

10-9. C. E. Brennan, Hydrodynamics of Pumps, Concepts ETl, Inc. and Oxford Uni-
versity Press, 1994.

10-10. Liquid Rocket Engines Turbopump Shafts and Couplings , NASA Space Vehicle
Design Monograph, NASA SP-8101, Sept. 1972.

10-11. Liquid Rocket Engines Turbopumps Gears, NASA Space Vehicle Design Mono-
graph, NASA SP-8100, March 1974.

10-12. Y. V. Demyanenko, A. I. Dimitrenko, and 1. 1. Kalatin, “Experience of Developing
Propulsion Rocket Engine Feed Systems Using Boost Turbopump Units,” AlAA
Paper 2003-5072, 2003.

10-13. Liquid Rocket Engine Turbopump Bearings , NASA Space Vehicle Design Mono-
graph, NASA SP-8048, March 1971.

10-14. Liquid Rocket Engine Turbopump Rotating Shaft Seals . NASA SP-8121, February
1978.

10-15. J. Kurokawa, K. Kamijo, and T. Shimura, “Axial Thrust Analysis on LOX-Pump,”
AlAA Paper 91-2410, June 1991.

10-16. M. C. Ek, “Solving Synchronous Whirl in High Pressure Turbine Machinery of
the Space Shuttle Main Engine,” Journal of Spacecraft and Rockets, Vol. 17,
No. 3, May-June 1980, pp. 208-218.

10-17. R. S. Ruggeri and R. D. Moore, Method for Prediction of Pump Cavitation Per-
formance for Various Liquids, Liquid Temperatures, and Rotating Speeds, NASA
TN D5292, June 1969.

10-18. Liquid Rocket Engine Turbopump Inducers, NASA Space Vehicle Design Mono-
graph, NASA SP-8052, May 1971.

10-19. T. Shimura and K. Kamijo, “Dynamic Response of the LE-5 Rocket Engine
Oxygen Pump,” Journal of Spacecraft and Rockets, Vol. 22, No. 2, March- April,
1985.

10-20. Liquid Rocket Engine Turbines, NASA Space Vehicle Design Criteria Monograph,
NASA SP-8110, January 1974.

10-21. S. Andersson and S. Trollheden, “Aerodynamic Design and Development of a
Two-Stage Supersonic Turbine for Rocket Engines,” AlAA Paper 99-2192, 1999.

10-22. D. Guichard and A. DuTetre, “Powder Metallurgy Applied to Impellers of Vinci
Turbopump,” International Symposium for Space Transportation of the XXI Cen-
tury, in CD ROM of the symposium. May 2003.

CHAPTER 11

ENGINE SYSTEMS, CONTROLS,
AND INTEGRATION

This chapter discusses the propellant budget, performance of complete or mul-
tiple rocket propulsion systems, the design of liquid propellant rocket engines
with pressurized or turbopump feed systems, engine controls, engine calibration,
system integration, and system optimization. Some of the content also applies to
solid propellant motors or hybrid propulsion systems.

11.1. PROPELLANT BUDGET

In all liquid propellant rocket engines the amount of propellant put into the
vehicle tanks is always a little greater than the nominal amount of propellant
needed to accomplish the intended mission. The extra propellant is needed for
uses other than providing thrust (e.g., residual propellant or valve actuation), to
compensate for changes from engine to engine, such as dimensional tolerances
causing slight changes in flow, for uncertainties of the construction of the engine
and minor changes in the flight plan. A propellant budget is the sum of all the
propellant utilization categories and losses in an engine; 11 are listed below. See
Ref. 11-1. The budget helps to determine how much propellant has to be loaded.
It is the aim to minimize this amount of propellant.

1. Enough propellant has to be available for achieving the required vehicle
velocity increase and/or the nominal attitude control maneuvers of the par-
ticular application and the particular flight vehicle or stage. The nominal
velocity increment is usually defined by systems analysis and mission opti-
mization using an iterative calculation based on Eq. 4-19 or 4-35. If there
are alternative flight paths or missions for the same vehicle, the mission

399

400 ENGINE SYSTEMS, CONTROLS, AND INTEGRATION

with an unfavorable flight path, such as higher drag or different orbit, and
the highest total impulse should be selected. This mission-required propel-
lant is usually the largest portion of the total propellants loaded into the
vehicle tanks.

2. In a turbopump system using a gas generator cycle, a small portion of the
overall propellant is burned in a separate gas generator. It has a lower
flame temperature than the thrust chamber gas and operates at a different
mixture ratio; this causes a slight change in the overall mixture ratio of
propellants flowing from the tanks, as shown by Eqs. 11-3 and 11-5.

3. In a rocket propulsion system with a thrust vector control (TVC) system,
such as a swiveling thrust chamber or nozzle, the thrust vector will be
rotated by a few degrees. Thrust vector control systems are described in
Chapter 18. There is a slight decrease in the axial thrust and that reduces
the vehicle velocity increment in item 1. The extra propellant needed to
compensate for the small velocity reduction can be determined from the
mission requirements and TVC duty cycle. It could be between 0.1 and
4% of the total propellant depending on the average angle position of this
thrust chamber.

4. In some engines a small portion of cryogenic propellants is heated, vapor-
ized, and the gas used to pressurize cryogenic propellant tanks. A heat
exchanger is used to heat liquid oxygen from the pump discharge and pres-
surize the oxygen tank, as shown schematically in Fig. 1-4. This method
is used in the hydrogen and oxygen tanks of the Space Shuttle external
tank (see Ref. 6-9).

5. Auxiliary rocket engines that provide for trajectory corrections, station
keeping, maneuvers , or attitude control usually have a series of small
restartable thrusters (see Chapter 4). The propellants for these auxiliary
thrusters have to be included in the propellant budget if they are supplied
from the same feed system and tanks as the larger rocket engine. Depend-
ing on the mission, the duty cycle, and the propulsion system concept,
this auxiliary propulsion system can consume a signiflcant portion of the
available propellants.

6. The residual propellant that clings to tank walls or remains trapped in
valves, pipes, injector passages, or cooling passages is unavailable for pro-
ducing thrust. It is typically 0.5 to 2% of the total propellant load. All
unused residual propellant increases the flnal vehicle mass at thrust termi-
nation and reduces the flnal vehicle velocity slightly.

7. A loading uncertainty exists due to variations in tank volume or changes
in propellant density or liquid level in the tank. This is typically 0.25 to
0.75% of the total propellant. It depends, in part, on the accuracy of the
method of measuring the propellant mass during loading (weighing the
vehicle, flow meters, level gauges, etc.).

8. The off-nominal rocket performance is due to variations in the manufacture
of hardware from one engine to another (such as slightly different pressure

1 1 .2. PERFORMANCE OF COMPLETE OR MULTIPLE ROCKET PROPULSION SYSTEMS 401

losses in a cooling jacket, in injectors and valves, or somewhat different
pump characteristics); these cause slight changes in combustion behavior,
mixture ratio, or specific impulse. If there are slight variations in mixture
ratio, one of the two liquid propellants will be consumed fully and an
unusable residue will remain in the other propellant’s tank. If a minimum
total impulse requirement has to be met, extra propellant has to be tanked to
allow for these mixture ratio variations. This can amount to up to perhaps
2.0% for each of the propellants.

9. Operational factors can result in additional propellant requirements, such
as filling more propellant than needed into a tank or incorrectly adjusting
regulators or control valves. It can also include the effect of changes in
flight acceleration from the nominal value. For an engine that has been
carefully calibrated and tested, this factor can be small, usually between
0.1 and 1.0%.

10. When using cryogenic propellants, an allowance for evaporation and cool-
ing down has to be included. It is the mass of extra propellant that is
allowed to evaporate (and be vented overboard while the vehicle is waiting
to be launched) or that is fed through the engine to cool it down, before the
remaining propellant in the tank becomes less than the minimum needed
for the flight mission. Its quantity depends on the amount of time between
topping off (partial refilling) of the tank and the engine start.

1 1 . Finally, an overall contingency or ignorance factor is needed to allow for
unforeseen propellant needs or inadequate or uncertain estimates of any
of the items above. This can also include allowances for vehicle drag
uncertainties, variations in the guidance and control system, wind, or leaks.

Only some of the items above provide axial thrust (items 1, 2, and sometimes
also 3 and 5), but all the items need to be considered in determining the total
propellant mass and tank volume.

The example shown in Table 11-1 is for a spacecraft pressure-fed engine
system, where the majority of the propellant is consumed in a larger axial thrust
chamber, and the second largest amount of propellant is fed to a set of several
small thrusters used for extensive attitude control maneuvers. There will be flights
where the mission may be more demanding or the engine performance may be
slightly lower, and extra propellant will be needed to accomplish the mission.
Conversely if the engine performance is actually slightly better than nominal
or where the mission can be accomplished with less total impulse (operate with
fewer or lower orbits), then the engine will consume less than the nominal amount
of propellant.

11.2. PERFORMANCE OF COMPLETE OR MULTIPLE ROCKET
PROPULSION SYSTEMS

The simplified relations that follow give the basic method for determining the
overall specihc impulse, the total propellant flow, and the overall mixture ratio

402 ENGINE SYSTEMS, CONTROLS, AND INTEGRATION

TABLE 11-1. Example of a Propellant Budget for a Spacecraft Propulsion System
with a Pressurized Monopropellant Feed System

Budget Element Typical Value

1 . Main thrust chamber (increasing the velocity
of stage or vehicle)

2. Flight control function (for reaction control
thrusters and flight stability)

3. Residual propellant (trapped in valves, lines,
tanks, etc.)

4. Loading uncertainty

5. Allowance for off-nominal performance

6. Allowance for off-nominal operations

7. Mission margin (reserve for first two items
above)

8. Contingency

85-95% (determined from mission
analysis and system engineering)
5-10% (determined by control
requirements)

0.5-2% of total load

0.5% of total load
0.5 -1.0% of total load
0.25-1.0% of total load
3-5% of items 1 and 2

1-5% of total load

Source: Adapted from data supplied by predecessor of Propulsion Products Center, Northrop Grum-
man Corporation

as a function of the corresponding component performance terms for complete
rocket engine systems. This applies to engine systems consisting of one or more
thrust chambers, auxiliaries, gas generators, turbines, and evaporative propellant
pressurization systems all operating at the same time.

Refer to Eqs. 2-5 and 6-1 for the specific impulse /j, propellant flow rate w
or m and mixture ratio r. The overall thrust Fga is the sum of all the thrusts from
thrust chambers and turbine exhausts and the overall flow tit is the sum of their
flows as shown in Eqs. 2-24 and 2-25. The subscripts oa, o, and / designate
the overall engine system, the oxidizer, and the fuel, respectively. Then repeating
these equations

(/)oa —

W„

T.F

goT,Fi

(11-1)

or fkoa = m

(11-2)

_J2Fio

(11-3)

Ew'f

These same equations should be used for determining the overall performance
when more than one rocket engine is contained in a vehicle propulsion system and
they are operating simultaneously. They also apply to multiple solid propellant
rocket motors and combinations of liquid propellant rocket engines and solid
propellant rocket booster motors, as in the Space Shuttle (see Eig. 1-14). All the
nozzles point in the same direction in Eqs. 2-24, 11-1, and 11-4.

Example 11-1. For an engine system (LOX/kerosene) with a gas generator similar to
the one shown in Fig. 1-4, determine a set of equations that will express (1) the overall

11.3. ENGINE DESIGN 403

engine performance and (2) the overall mixture ratio of the propellant flows from the tanks.
Let the following subscripts be used: c, thrust chamber; gg, gas generator; and tp, tank
pressurization. For a nominal burning time f, a 1% residual propellant, and a 6% overall
reserve factor, give a formula for the amount of fuel and oxidizer propellant required
with constant propellant flow. Ignore stop and start transients, thrust vector control, and
evaporation losses.

SOLUTION. Only the oxidizer tank is pressurized by vaporized propellant. Although
this pressurizing propellant must be considered in determining the overall mixture ratio,
it should not be considered in determining the overall specific impulse since it stays with
the vehicle and is not exhausted overboard.

(f.v)oa

^ oa

Fc + Fgg

{me + mgg)go

{FIq^C “t“ (hlo)gg “t“ {^o)tp

(mf)c + (mf)gg

(11-4)

(11-5)

nif = [{mf)c + (m/)gg]?(1.00 + 0.01 + 0.06)

nio = [{mo)c + (ffio)gg + («L )?;,]? (1-00 + 0.01 + 0.06)

For this gas generator cycle the engine mixture ratio or Vga is different from the
thrust chamber mixture ratio rc = {m„)c/{mf)c. Similarly, the overall engine specific
impulse is slightly lower than the thrust chamber specific impulse L = Fe/ihe. However,
for an expander cycle or a staged combustion cycle these two mixture ratios and two
specific impulses are the same, provided that there are no gasified propellanf used for
tank pressurization. Engine cycles are explained in Section 6.6.

The overall engine specific impulse is influenced by the propellants, the nozzle
area ratio, and the chamber pressure, and to a lesser extent by the engine cycle,
and the mixture ratio. Table 11-2 describes 10 different rocket engines using
liquid oxygen and liquid hydrogen propellants designed by different companies
in different countries, and shows the sensitivity of the specific impulse to these
parameters. References 11-2 to 11-4 give additional data on several of these
engines.

11.3. ENGINE DESIGN

The approach, methods, and resources used for rocket engine preliminary design
and final design are usually different for each design organization and for each
major type of engine. They also differ by the degree of novelty.

1. A totally new engine with new major components and some novel design
concepts will result in an optimum engine design for a given application,
but it is usually the most expensive and longest development approach.
One of the major development costs is usually in sufficient testing of

TABLE 11-2. Comparison of Rocket Engines Using Liquid Oxygen and Liquid Hydrogen Propellants'

bX)

c/3

'5b

»n

£:•

■o

n «

>§ < ai.

*'1

0 )

S-I

o’

l-H

c/3

C/3

1-1

o o q o
in in »n

CNJ

»n

i>;

^ CM
»n ^

_ 00
O CM

»n

s 5

s ^

^ o ^

'^'3 3

0 3 3

Oh Oh 3

C/5 S d

''t

^23

»n

^ O

''t

<N §

'55l-

cm"

CM*

»n

r-;

1—

r- ^

1-H

no"

'55l-

C--

C 3 3
O- 4 =
00 C/5

D .

-Q

•< (M •<

c/3

‘C

c/3

M (D

OX)

O "S
t— ) m

W §

^ o

a o ;

§ •

Q S

PJ u

3
3

- "2
3 Q
O 4^
'z: o
cfl O

== c2

40 ^

3 >>

o S

-3 2

bJ) >
3

bX)

biD

oo"

8 £

D *0

"2 X5

2^ D

O "O
O 3
V 3
^ Oh

g w
> »o

' o ^

' C> (N

' T5
; ^ ON

^ 00
. 3 ^

c«

3
3

C-’' ^
PP ^

c/5

1 -H

3
<1^ ^
OX Co

M ON
3 ON
bfl ^

e ^
3 O

c/5

sP'

^ s

C ^

g 00

bX) 2

Cfl ^

M 8

C 4

1 -H

2 ?
^ PU
>H

'3 ; 2 T

Vh on

ci”;

«i E
5 ^ O

Oh ^

404

Additional information on most of these engines is in this book. Use the index to find it.

11.3. ENGINE DESIGN 405

components and several engines (under various environmental and perfor-
mance limit conditions), in order to establish credible reliability data with
enough conhdence to allow the initial flights and initial production. Since
the state of the art is relatively mature today, the design and development
of a truly novel engine does not happen very often.

2. New engine using some major components or somewhat modified key com-
ponents from proven existing engines. This is a common approach today.
The design of such an engine requires working within the capability and
limits of existing or slightly modified components. It requires less testing
for proving reliability.

3. Uprated or improved version of an existing, proven engine. This approach
is quite similar to the second. It is needed when an installed engine for
a given mission requires more payload (which really means higher thrust)
and/or longer burning duration (more total impulse). Uprating often means
more propellant (larger tanks), higher propellant flows and higher chamber
and feed pressures, and more feed system power. The engine usually has
an increased inert engine mass (thicker walls).

In a simplified way, we describe here a typical process for designing an engine.
Chapter 19 and Refs. 11-5 and 11-6 describe this process and the selection of the
propulsion system from a different point of view. At first the basic function and
requirements of the new engine must be established. These engine requirements
are derived from the vehicle mission and vehicle requirements, usually deter-
mined by the customer and/or the vehicle designers, often in cooperation with
one or more engine designers. The engine requirements can include key param-
eters such as thrust level, the desired thrust-time variation, restart or pulsing,
altitude flight profile, duty cycle, maximum accelerations, engine locations within
the vehicle, and limitations or restraints on cost, engine envelope, test location, or
schedule. It also includes some of the factors listed later in Table 19-5. If an exist-
ing proven engine can be adapted to these requirements, the subsequent design
process will be simpler and quite different than the design of a truly new engine.

Usually some early tentative decisions about the engine are made, such as the
selection of the propellants, their mixture ratio, or the cooling approach for the
hot components. They are based on mission requirements, customer preferences,
past experiences, some analysis, and the judgment of the key decision makers.
After some studies additional selection decisions can also be made, such as having
one, two, or more thrust chambers fed from the same feed system, redundancy
of auxiliary thrusters, or type of ignition system.

A systematic approach, a good set of analyses, the use of system engineering,
good coordination with the customers, key vendors, and vehicle designers — all
are needed for a good preliminary and final design. Before a meaningful proposal
for an engine can be prepared, a preliminary design has to be completed. See
Refs. 11-5 to 11-7. One of the early design decisions is the choice of feed
system: pressurized gas feed or pump feed. The next two paragraphs give some
guidelines.

406 ENGINE SYSTEMS, CONTROLS, AND INTEGRATION

A pressurized feed system (see Fig. 1-3) usually gives better vehicle perfor-
mance for low values of total impulse (thrust less than about 4.5 kN or 1000 Ibf
with up to perhaps 2 min duration). A pump feed system gives better vehicle
performance for high thrust (say above 50,000 Ibf or approximately 222 kN) and
a long cumulative duration — more than a couple of minutes. For intermediate
values of total impulse the choice can go either way, and it is not easy to make
the decision only on the basis of total impulse. If the chamber pressure of a
pressurized feed system is relatively high (say about 2.4 to 3.5 MPa or about 350
to 500 psia and occasionally more), then the inert weight of the thrust chamber
will be high, but the thmst chamber will be small and can usually ht into an
engine compartment. The vehicle propellant tanks and pressurizing gas tank will
be at relatively high pressure and will be heavy. For a relatively low chamber
pressure (0.689 to 1.379 MPa or 100 to 200 psia) the vehicle tank pressure will
be lower and the tank walls thinner, but the thrust chamber will be very large
in size and often will exceed the limits of the engine compartment, unless it
has a low nozzle area ratio, which implies lower performance. Even in some
cases of low chamber pressure, a pressurized feed system can be heavier than
a comparable engine with a pump feed system. Pressurized feed systems have
been relatively simple, very reliable, and they allow fast starts and fast restarts.
With restarts the cooling of the thrust chamber can become a problem. Because
of the proven reliability of pressure feed system, NASA has at times conser-
vatively selected pressurized feed systems for certain space applications, such
as the Apollo service module engine (21,900 Ibf thrust), even though there is a
major weight penalty and a somewhat inferior vehicle performance compared to
a pump-fed system of equal total impulse. Also a decision has to be made on
using either a pressurized system with a gas pressure regulator or alternatively a
blow-down system. See Table 6-3 and Section 6.4.

For a turbopump-fed liquid propellant rocket engine (see Fig. 1 -4) the overall
inert weight of propellant tanks and engine will be considerably lighter and
usually the vehicle performance will be somewhat better. It usually operates
at high chamber pressures (3.5 to 24.1 MPa or about 500 to 3500 psia), the
thrust chamber is not normally protrading from the vehicle. The smaller and
shorter thrust chamber often allows a shortening of the vehicle with a savings
in vehicle structure. This further improves the vehicle’s performance modestly
and the higher specific impulse will slightly reduce the amount of propellant
needed for the mission. Compared to an engine with a pressurized feed system,
the savings in inert mass (thin vehicle tank walls) and less propellant will allow
a smaller, lighter, and probably lower cost vehicle with a somewhat superior
performance. The engine with a TP is more complex with more parts and the
engine will generally be heavier; however, the vehicles propellant and gas tanks
will be much lighter and they will more than compensate for the heavier engine.
It will take more tests and more effort to prove high reliability in an engine with
a pump feed system. At the higher chamber pressures the heat transfer will be
higher and cooling can become a problem, but heat transfer has been solved in
other earlier high-pressure rocket engines. Also restart will be more complex. One

11.3. ENGINE DESIGN 407

of several engine cycles has to be selected. High reliability has been achieved in
many turbopump-fed large rocket engines.

Trade-off studies between several options are appropriate at this time. With a
modified existing engine these parameters are well established and require fewer
trade-off studies or analyses. Initial analyses of the pressure balances, power
distribution between pumps and turbines, gas generator flow, propellant flows
and reserves, or the maximum cooling capacity are appropriate. Sketches and
preliminary estimates of inert mass of key components need to be made, such as
tanks, thrust chambers, turbopumps, feed and pressurization systems, thrust vector
control, or support structure. Alternate arrangements of components (layouts) are
usually examined, often to get the most compact configuration or control of the
travel of the center of gravity. An initial evaluation of combustion stability, stress
analysis of critical components, water hammer, engine performance at some off-
design conditions, safety features, testing requirements, cost, and schedule are
often performed at this time. Participation of appropriate experts from the field of
manufacturing, field service, materials, stress analysis, or safety can be critical for
selecting

Featured

Top

Featured

Top

Featured

Top

Featured

Top

Featured

Top

Full text of "Rocket propulsion elements"

See other formats